JP4913992B2 - Vehicle travel support system - Google Patents

Description

  The present invention relates to a technique for performing vehicle travel support (including lighting control of a lamp) while predicting the shape of a travel path of the host vehicle.

  The vehicle travel support system includes, for example, a system that acquires road linear data from a navigation device and performs automatic transmission control (ATC), or a traction control system (TSC) or a stability control system. (VSC: Vehicle Stability Control system) is known, and the latter two can obtain the optimal driving force so that the tire does not slip or the driving wheel spins, etc. when driving on a curved road. The purpose is to increase the stability when traveling or on a slippery road surface. In addition, a system has been developed that warns when a car enters a curved road.

  With regard to a support system capable of controlling the light distribution of the vehicle headlamps corresponding to the own vehicle traveling road, for example, based on the map information and the own vehicle position information on the map, up to the turning point on the road ahead of the own vehicle There is known a system that controls an irradiation range and the like while judging a distance and a bending direction (see Patent Document 1).

  By the way, it is accuracy that is a problem when performing curved road determination or road linear prediction on a traveling road ahead of the host vehicle. For example, the following method can be cited.

(I) A method for obtaining a density state of nodes (road notation points on the road map database) existing on the own vehicle traveling path (nodes are dense on curved roads and coarse on straight roads).
(II) Method of obtaining radius of curvature of runway by circular interpolation from position data of three nodes existing ahead of current vehicle position (III) Change of normal vector at node point based on calculation result of (II) above (IV) A method of calculating a road linear characteristic by defining a link vector connecting nodes existing on the own vehicle traveling path and calculating an outer product between adjacent vectors (V) ) A method of calculating the road alignment by interpolating a plurality of nodes existing on the own vehicle traveling path with a spline curve or a free curve.

JP-A-2-296550

  However, it is necessary to consider the accuracy in applying the conventional system. For example, the requirement for accuracy becomes severe in application to a visual support system such as a light distribution control type vehicle headlamp device. .

  Among the above methods, the methods (I) to (IV) are applicable to automatic transmission control, traction control, stability control, etc., although there is a difference in accuracy. The reason is mainly due to the fact that the influence of control deviation or the like is reflected in the driver's experience, and the accuracy required for the control itself is lower than in the case of optical control.

  In order to improve the accuracy, the method (V) is preferable. However, the calculation amount necessary for calculation of road alignment is increased, and spline curves or the like are applied depending on the number and arrangement of node points according to the road alignment. Is inappropriate (for example, in road shapes where slalom and curved roads are combined, if the number of node points and the order of the interpolation function are not appropriate, irregular amplitudes occur and the interpolation curve undulates) It is necessary to deal with the above.

  In the application to the driving support system, higher accuracy is required in the form based on the driver's visual recognition, such as the light distribution control of the headlamps, than in the form based on the driver's experience and voice notification. That is, if the road alignment is misjudged due to insufficient accuracy, irradiation control that does not match the actual situation is performed, which not only hinders driving but also affects road users and the like.

  For example, when the method (II) is applied to calculate the road linear radius in front of the vehicle by three-point circular interpolation and the headlamp light distribution is controlled based on the result, the following method can be cited.

(II-1) A method of predicting the vehicle arrival position after a predetermined time of the host vehicle and controlling the direction of the irradiation beam with the position as the gazing point. (II-2) The irradiation beam with the clipping point of the curve as the gazing point. How to control the direction.

  When three-point circular interpolation is applied in these controls, for example, if a phenomenon occurs in which the direction of the irradiation beam changes instantaneously (turning the beam) while traveling on a curved road, the driver's field of view is of course opposite. There is concern about the impact on car drivers.

  FIG. 12 shows an example of a road line shape connecting the node points N1 to N4, and shows an S-shaped curved road.

  The above phenomenon occurs between the radius R1 calculated at the three nodes N1, N2, and N3 based on the current vehicle position and the radius R2 calculated at the next three nodes N2, N3, and N4. Is caused by the fact that the road turning direction is reversed when passing through the node N2 in this example. That is, it occurs when attempting to illuminate the vehicle predicted position or the clipping point after a predetermined time has elapsed, and it is known that the above (II-1) is remarkable. This phenomenon also occurs when the map matching is poor and the current position is misidentified or the position error is large.

  As described above, if the shape characteristics of the road alignment are not grasped with sufficient accuracy, problems such as control misalignment and delay in alarm timing may occur, causing discomfort and discomfort during driving. . In addition, as described above, when irradiation control is performed by applying 3-point circular interpolation to a dark cloud without considering the presence of inflection points, etc. in the road alignment, it affects the most sensitive visual sense among the five senses. Is a problem.

  Therefore, an object of the present invention is to realize a control that does not cause a sense of incongruity to the driver by accurately predicting the road shape of the own vehicle traveling path in the vehicle travel support system.

  In order to solve the above-described problems, the present invention reads the position data of a plurality of nodes located before and after the current position of the vehicle using the current position data of the vehicle and the road map data, and connects adjacent nodes. The road shape characteristics (including linearity and curvature characteristics) are determined from each formed link, and the road alignment is estimated by connecting the plurality of nodes according to an interpolation process according to the determination result. The vehicle driving support device is configured to be controlled according to the predicted estimation result of the running road shape, and includes the following components.

  A vehicle current position detection unit that acquires current position data of the vehicle using satellite communication or road-vehicle communication.

  Node data position extraction means for reading position data of a plurality of nodes located before and after the current position of the vehicle using the current position data of the vehicle and road map data.

  Link characteristic determining means for determining the characteristics of each link formed by connecting adjacent nodes among a plurality of nodes and the type of the runway section.

  Road linearity generation means for generating road shape data modeled by connecting a plurality of nodes by interpolation processing.

Control means for sending a control output corresponding to the estimation result of the road shape data to the driving support device.
The type of the road segment determined by the link characteristic determination unit is any of a plurality of track segments, and the link characteristic determination unit determines the type of the road segment and the determination result is Send to road alignment generation means.
The road alignment generation unit performs an interpolation process according to the determination result determined by the link characteristic determination unit.

  Therefore, in the present invention, it is possible to accurately estimate the road shape by applying an interpolation process according to the shape characteristic of the road.

According to the present invention, for the road ahead of the host vehicle , the shape of the road alignment is accurately predicted by performing the interpolation process according to the type of the typed road section determined by the link characteristic determination unit , It is possible to realize driving support control that is less uncomfortable for the driver, and therefore, there is no significant increase in calculation amount or processing load. For example, it is suitable for an application that requires high accuracy in a form in which the influence of control deviation or the like is recognized visually by the driver, such as an irradiation control device for a vehicle headlamp.

  And by using the above-mentioned means, it is possible to calculate the road shape and the type of the runway section required for driving support of the vehicle accurately and in a short time, and for that purpose, a high processing capacity is required. Absent.

  Further, in the configuration form in which the link characteristic determining unit determines any one of the inflection section, the road section, and the straight section including the inflection point of the road shape, the road section can be classified and discriminated.

  In order to be able to easily determine the type of the road section, for example, the link length between adjacent nodes is compared with a reference distance value determined in advance based on the second speed converted value of the legal travel speed, and the link When the length is equal to or greater than the reference distance value, the runway section related to the link is determined as a straight section, and when the link length is less than the reference distance value, the runway section related to the link is changed to an inflection section or a curve. A configuration form that discriminates from a road section is preferable.

  Then, in order to be able to accurately determine the type of the runway section, a configuration form in which the sign change of the outer product of the link vectors obtained as the difference between the link length between adjacent nodes and the position vector of the nodes is obtained. Alternatively, there is a configuration in which the radius of curvature obtained by performing circular interpolation on three nodes constituting two adjacent links and the sign change of the radius of curvature are obtained.

  In the embodiment provided with the vehicle current position correcting means for correcting the current position data of the own vehicle, the direction change in the current link is calculated based on the road line shape equation estimated from the road shape data generated by the road line shape generating means. Calculate the prediction range. Then, when the direction detection data of the own vehicle is out of the predicted range, it is determined that the current position of the own vehicle does not exist on the current link, and the target is changed to another link located around the current link. Continue to correct the current vehicle position. If the direction detection data of the host vehicle is included in the predicted range, a position that is the same direction as the direction detection data on the current link or closest to the data is estimated as the host vehicle current position. It ’s fine. This makes it possible to accurately grasp the current position of the host vehicle and is effective in improving accuracy.

  FIG. 1 shows an example of the basic configuration of the present invention.

  The vehicle travel support system 1 acquires the current position data of the vehicle, controls the travel support device mounted on the vehicle while estimating the road shape around the current position of the vehicle using the road map data. (The number in parentheses indicates a sign.)

・ Vehicle current position detection means (2)
・ Road information provision means (3)
Node data position extraction means (4)
・ Link characteristic discrimination means (5)
・ Road alignment generation means (6)
.Control means (7)
・ Running support device (8)
.Vehicle current position correction means (9)
.Direction detection means (10)
.Vehicle running state detection means (11)
-Vehicle steering state detection means (12).

  The own vehicle current position detecting means 2 is provided for obtaining current position information of the own vehicle, and for example, a navigation device using GPS (Global Positioning System) by artificial satellite communication or road-to-vehicle communication is used. The acquired own vehicle current position data is sent to the node data position extracting means 4.

  The road information providing means 3 is necessary for obtaining information on the travel route using a road map database or the like. For example, road map data provided from a recording medium or the like is sent to the node data position extracting means 4.

  The node data position extracting means 4 uses the current position data of the vehicle from the current position detection means 2 and the road map data of the road information providing means 3 to determine the positions of a plurality of nodes located before and after the current position of the own vehicle. Read data. The extracted node position data is sent to the link characteristic determination means 5 and the road alignment generation means 6 at the subsequent stage.

  The link characteristic discriminating means 5 determines the characteristics of each link formed by connecting adjacent nodes among a plurality of nodes related to the own vehicle traveling path, for example, the shape characteristics including linearity and curvature characteristics and the type of the traveling path section. The determination result is sent to the road alignment generation means 6.

  The road alignment generation means 6 is provided to generate road alignment shape data modeled by connecting a plurality of nodes related to the own vehicle travel path by interpolation processing, and the data is sent to the control means 7. The

  The control means 7 sends a control output corresponding to the estimation result of the road shape data relating to the travel path predicted from the current position of the vehicle to the travel support device 8. The driving support device 8 includes, for example, a vehicle headlight irradiation control device (light distribution control type headlight device, etc.), an automatic transmission control device, a vehicle stability control device, various alarm devices, and the like. Is mentioned.

  The vehicle current position correction means 9 is provided for correcting the vehicle current position data based on the road alignment data from the road alignment generation means 6 and the detection data from the vehicle direction detection means 10 (direction sensor such as a gyro sensor). Therefore, the control accuracy can be improved by the correction.

  The vehicle running state detection unit 11 detects the vehicle speed, acceleration, and the like of the host vehicle, and sends the detection result to the irradiation control unit 8. For example, the current vehicle speed data is required to calculate the travel time required to reach a curved road or intersection.

  The vehicle steering state detection unit 12 detects the steering state or the traveling direction (azimuth) of the host vehicle, and sends the detection result to the irradiation control unit 7. For example, the steering state of the vehicle including the turning direction of the vehicle is detected using a steering sensor, an angular velocity sensor, or the like.

  First, the definition of the nodes and links necessary for grasping the road alignment relating to the own vehicle traveling road will be described with reference to FIG.

  In the figure, “Ni” (i = 0, 1-6) represents the i-th node, and “Li” (i = 0, 1-5) represents the i-th link. In this example, a link formed by connecting two adjacent nodes Ni and Ni + 1 with a line segment is defined as Li (the length of Li, that is, the link length is not necessarily constant). ).

  The home base symbol shown in the figure represents the current position of the vehicle (the direction of the sharp point indicates the vehicle traveling direction), and the node N1 existing behind it is defined as the “current node” and in front of it. The located node N2 is defined as “front node”. As for the link, the link where the vehicle exists, that is, the link L1 connecting the nodes N1 and N2 is defined as the “current position link”, and the link L2 ahead is defined as the “forward link”.

  Information regarding a plurality of nodes positioned before and after the current vehicle position is sent to the node data position extraction unit 4 where position data is read and the like.

The link characteristic discriminating means 5 discriminates the shape characteristic of a predetermined link and grasps, for example, any one of a curved road section, an inflection section (including an inflection point), and a straight section.

  In a road map database used for a navigation device or the like, for example, the following characteristic properties are recognized between the node dot characteristics and the road linear characteristics.

-The node hitting interval is rough on straight roads (and therefore the link length is large), and the node hitting point interval is dense on curved roads.-The road class is low for the node hitting interval in the curved section (design vehicle speed value). The results are shown in the table below as examples of road types and node intervals (link lengths) of specific sections.

  Note that the node spacing in the above table is shown as a multiple of a value obtained by converting the road design speed into a second speed. Further, the inflection section in the above table includes a curved road. As for the road type, it can be determined from the map data whether the own vehicle runway is an expressway, a national road, a prefectural road, a municipal road, or the like.

  When distinguishing sections according to road design speed or road class, it is possible to identify straight roads, curved roads, or intermediate roads (inflection sections, curved roads, etc.) using node intervals. It is.

As described above, since there is a problem in applying the three-point circular interpolation to the road alignment including the inflection point, the predicted runway is defined as a straight section or a curved section in consideration of the link connection relation. Then, after dividing into inflection sections, an interpolation process according to the classification is adopted. This simplifies the equation relating to road alignment and simplifies the road alignment generation logic (however, when a runway is divided into a plurality of sections, for example, for a node existing in a predetermined section, a curve start (Note that there will be a need to define new points, turn points, etc., or reset them.)

  Examples of the section discrimination include the following methods, which can be used independently or in combination.

(I) A configuration in which a link length is compared with a reference distance determined from a legal traveling speed of the travel route to determine a travel route section related to the link. (II) An outer product of the link length and the link vector is used to relate to the link. Configuration form for discriminating the runway section (III) Configuration for discriminating the runway section related to the link based on the sign change of the radius R calculated by circular interpolation and the R value (absolute value) using three adjacent nodes. Form.

  First, in (I), for example, the link length between adjacent nodes is compared with a reference distance value determined in advance based on the second speed related to the legal travel speed. The reference distance value can be defined as a constant multiple of the second speed converted value, taking Table 1 as an example. For example, at a speed of 80 km / hr, when a reference distance value equivalent to 10 times the second speed value is defined, if the link length is equal to or greater than the reference distance value, the runway section related to the link is determined as a straight section. The Further, when the link length is less than 10 times the second speed value, the runway section for the link to be determined is an inflection section (5 to 10 times) or a curved section (5 times or less). Is determined. In this configuration, since it is only necessary to compare the link length with the reference distance, it is preferable to combine with (II) and (III) in terms of accuracy.

  In the above (II), by determining the sign change of the outer product of the link vectors obtained as the difference between the link length between adjacent nodes and the position vector of the nodes, the type of the runway section related to the target link can be determined. . That is, the outer product of the link vector corresponding to the link “Li-1” and the link vector corresponding to the link “Li” is denoted as | Li−1 × Li |, the link vector corresponding to the link “Li”, and the link When an outer product with a link vector corresponding to “Li + 1” is denoted as | Li × Li + 1 |, it is a section over 3 links (4 nodes) by grasping how their signs change. Judgment is possible. For example, each cross product value is zero in the straight section, and the sign of the cross product changes from positive to negative (or from negative to positive) in the inflection section.

  In (III) above, it is possible to know the curvature radii Ri-1, Ri, Ri + 1,... By performing circular interpolation on the three nodes constituting the two adjacent links. By obtaining the absolute value, the type of the runway section related to the target link can be determined. For example, when the radius of curvature is greater than or equal to a predetermined reference value (such as 1000 m), the straight section is determined, and the curvature radius is less than the predetermined reference value, and the sign of the curvature radius is positive to negative. When changing to (or from negative to positive), an inflection section or the like is determined.

  In (II) and (III), it is possible to accurately determine the type of the track section using the vector product and the radius of curvature.

  Next, the interpolation process in the road alignment generation means 6 will be described.

  The interpolation method used is, for example, as shown below.

-Three-point circular interpolation method-Bisection circle method-Newton's four-point forward interpolation method.

  Note that it is preferable to employ a smoothing process described later in a section where the use of these methods is not appropriate.

  FIG. 3 is an explanatory diagram of the three-point circular interpolation method.

  Vertical bisectors m1 and m2 are drawn with respect to the adjacent links L1 and L2, respectively, as indicated by broken lines, and the intersection “O” between them is set as the center of the arc (center of curvature).

  When the arc angle with respect to L1 is denoted as “α” and the arc angle with respect to L2 is denoted as “β”, the radius “R” of the arc extending over the links L1 and L2, that is, the arc passing through the nodes N1 to N3 is I want to.

  When three-point interpolation is used continuously, for example, an arc having a radius Ri-1 is obtained from the adjacent link pair "Li-1, Li", and the next link pair "Li, Li + 1" is obtained. When the arc having the radius Ri is obtained, there is a problem that both arcs are partially overlapped on the nodes Ni to Ni + 1. That is, the discontinuity of the runway shape becomes significant in the inflection section where the sign of the radius changes.

  Therefore, the bisected circle method shown in FIG. 4 is used.

  In this case, circular interpolation is performed for the first three points, and the tangent circle on the extension line of the bisector is used from the next node.

  An arc of Ri-1 = Ri is obtained by three-point circular interpolation for the nodes Ni-1, Ni, Ni + 1 shown in the figure, and the next link Li + 1 has an arc (radius Ri + 1). The tangent circle is defined so that the center position is located on the line segment connecting the arc center having the radius Ri and the node Ni + 1. That is, in the arc having the radius Ri related to the previous link Li, the center of the arc (radius Ri + 1) is positioned on the normal line at the node Ni + 1, and the arc is one of the tangent circles at Ni + 1. Set to be part.

  When the arc angle for the link Li is “αi”, the arc angle for the link Li + 1 is “αi + 1”, and the angle between the links Li and Li + 1 is “θi”, the relationship shown in the following equation is established. To establish.

  The outer product of link vectors (vector product of links) is calculated by “| Li × Li + 1 | = | Li ||| Li + 1 | · sin (θi)”, and | Li × Li + 1 | When the sign changes between | Li + 1 × Li + 2 |, it can be seen that the link Li + 1 is an inflection section.

  In the case of an inflection section, as is clear from the equation [2], it is necessary to redefine the track alignment using three-point circular interpolation or the like under the condition of “θi <(αi / 2)”. is there.

  FIG. 5 is an explanatory diagram of Newton's four-point forward interpolation method.

  In this figure, the horizontal axis is the x-axis and the vertical axis is the y-axis, showing the points Ni-2, Ni-1, Ni, Ni + 1, Ni + 2, Ni + 3, etc. (A broken line indicates a polygonal line connecting the node points by linear interpolation.) “Pi-1” and “Pi + 2” indicate node points after rearrangement.

  The current position p (x, y) and direction of the vehicle are indicated by a home base symbol, Ni is the current node, and the link connecting Ni and Ni + 1 is the current position link.

  In the method of interpolating each node with a spline curve, considering the fact that the calculation amount is large and the processing time is long, a form in which a cubic polynomial is derived using four nodes is preferable.

  Since Newton's forward interpolation method requires data obtained at equal intervals in the x-axis direction, for example, when the interval (x interval) on the x-axis between Ni and Ni + 1 is expressed as “Δω”. The nodes are rearranged at Δω as indicated by x0, x1, x2, and x3 in the figure. Here, “x1 = x0 + Δω”, “x2 = x0 + 2 · Δω”, and “x3 = x0 + 3 · Δω”.

  Of the four nodes, Ni-1 and Ni + 2 are subject to rearrangement, and Pi-1 and Pi + 2 are selected. That is, when the x coordinate of Ni is x1 and the x coordinate of Ni + 1 is x2, the vehicle travels so that the x coordinate of Pi-1 is x0 and the x coordinate of Pi + 2 is x3. Rearrangement is performed on Pi-1 located behind Ni-1 and Pi + 2 located ahead of Ni + 2 in the direction. In the calculation of the rearrangement points Pi-1 and Pi + 2, in this example, linear interpolation is adopted, and points are set on broken lines indicated by broken lines (Ni-2 and Ni-). Pi-1 is located at the intersection of the line connecting 1 and the straight line “x = x0”, and the intersection of the line connecting Ni + 2 and Ni + 3 and the straight line “x = x3” However, the present invention is not limited to this, and the rearrangement point may be calculated using the property of cubic interpolation itself.

  When each coordinate of four points that are equally spaced (Δω) on the x-axis is expressed as “(xi, yi)” (i = 0, 1, 2, 3), the following formula is used using Newton's forward formula: A cubic equation “y (x)” shown is obtained.

Incidentally, "Δy0" is first-order difference, "delta ² y0" the second floor difference "delta ³ y0" is a third-order difference, is defined by the following equation.

  In the relationship with FIG. 5, (x0, y0) is the coordinate of Pi-1, (x1, y1) is the coordinate of Ni, (x2, y2) is the coordinate of Ni + 1, and (x3, y3) is Pi. Each +2 coordinate is shown.

  FIG. 6 is an explanatory diagram of the smoothing process.

  As shown in L1 and L2, in a section (bending section) in which a section having a long link length is connected in a polygonal line, a process of smoothly connecting links with an arc having a predetermined radius R in the vicinity of the bending point indicated by the node N2 is performed. Is preferred.

  “Θ” in the figure indicates an angle formed between the links L1 and L2. When the angle of the bending section is expressed by using this θ, it is desirable to set the radius R value to a value slightly larger than the design radius (linear radius) of the road alignment (depending on the θ value) ( For example, in the case of a road at a speed of 40 km / hr, R is set to about 30 m, and the gap distance between the bending point and the arc (R) is set so as not to exceed one lane width.

  In the section subjected to the smoothing process, it is preferable that a new node is generated and the runway shape is defined in detail as indicated by points Na and Nb. When the arc distance between the apex of the arc (R) near the new node and the bending point is denoted as “d”, it is represented by “d = R · (θ / 2)”.

  FIG. 7 shows a definition example of the runway shape in a tabular form, and illustrates a combination of a curve section, an inflection section, and a straight section for the current position link L1 and the front link L2.

  (A) The figure shows a case where both L1 and L2 are curved sections, and illustrates a slalom section and a hairpin curve section. In L0, L1, and L2, a three-point circular interpolation method or a bisection circle method is preferable.

  (B) In the figure, L1 is an inflection section and L2 is a track section, for example, an entry section to a curve path or a slalom section. In this case, four-point forward interpolation is preferable for L0, L1, and L2, but it is also possible to define L1 as a straight section in sections other than L1 and perform three-point circular interpolation in other sections.

  (C) In the figure, L1 is a straight section and L2 is a curved section. In the illustrated example, L0 and L1 are straight sections (thus, linear linear interpolation), and N2 can be defined as a curve start point.

  In FIG. 4D, L1 is a curve section and L2 is an inflection section. In L1, L2, and L3, four-point forward interpolation is preferable with the current position link as the center. In a section other than L2, L2 may be a straight section and three-point circular interpolation or the like may be performed.

  (E) In the figure, both L1 and L2 are inflection sections. Although it is preferable to apply a four-point forward interpolation method centered on the current position link in L0, L1, and L2, smoothing processing may be performed by defining a bending path with N2 as a bending point.

  (F) In the figure, L1 is a straight section and L2 is an inflection section. In a bending section including L1 and L2, N2 may be defined as a curve starting point, and L2 and L3 may be defined as a curve section.

  (G) In the figure, L1 is a curved section and L2 is a straight section (when the link length is long). In a section other than L2, 3-point circular interpolation or 4-point forward interpolation is desirable. Alternatively, the runway may be defined with N2 as the track end point and N3 as the track start point.

  (H) In the figure, L1 is an inflection section and L2 is a straight section (when the link length is long). In a section other than L2, 3-point circular interpolation or 4-point forward interpolation is desirable. Further, L1 and L2 may be defined as bent sections centered on N2.

  (I) In the figure, L1 is a straight section and L2 is a straight section (when the link length is long). In the bending section shown on the right side, N2 may be a bending point, and the smoothing process may be performed with a linear radius in road design.

  The definition example and interpolation method of the above runway shape are as shown in the table below.

  FIG. 8 exemplifies the shape of the runway, and the node points are indicated by circles.

  In part A in the figure, the front and rear of the curved section are inflection sections and curved sections, and therefore it is preferable to use the four-point forward interpolation method.

  The bisected circle method is used in the curved section shown in part B in the figure, and the smoothing process can be used in the bent part shown in part C in the figure.

  In the above example, 3 × 3 = 9 types of road linear shape definitions are possible, and an appropriate interpolation method can be used according to the determination result of the link characteristics and the travel section. Since it is difficult to deal with all road shapes as a real problem, it is preferable to appropriately use the smoothing process using the radius of curvature in the design of road alignment as described above (for example, an actual curved road is ideal) Since most of them have a curved appearance rather than a shape in which two straight lines are connected, there is a problem that the control timing is delayed without smoothing processing.

  The road alignment generation means 6 is preferably configured to calculate and output, for example, the following information based on the result of data generation related to the road shape.

・ Turning radius and starting point and end point of the road predicted ahead of the vehicle ・ Bending point direction angle (clipping point angle, ie vehicle (An angle between the line of sight, the contact point on the road alignment, and the line connecting the vehicle with respect to the axis extending in the traveling direction)
-Bending angle on a bending path (see θ in FIG. 6), etc.

  The data output from the road alignment generation means 6 is also used in the vehicle current position correction means 9 for correcting the current vehicle position data. For example, the vehicle current position correction process is performed according to the following procedure.

(1) Calculate the predicted range of azimuth change in the target link (link assumed to be the current position link) based on the road linear shape equation estimated from the road shape data generated by the section discrimination and interpolation processing described above (2) If the direction detection data related to the host vehicle is out of the predicted range of (1), it is determined that the current position of the host vehicle does not exist on the link of (1), and the vehicle is positioned around the link. Change the target to another link (link before and after) and continue the correction process of the current position of the vehicle (3) When the direction detection data related to the vehicle is included in the prediction range of (1), The position that is the same as the direction detection data on the link of (1) or that is the closest direction to the data on the current link is estimated as the current position of the host vehicle, and the position is corrected (4) (1) Return to and repeat the process.

In FIG. 9, the azimuth estimated from the current position candidate indicated by the point “Pe” on the road (see arrow y 1) and the azimuth actually detected by the GPS azimuth sensor (see arrow y 2) are different. In the case, according to the above (2) , a situation is shown in which the correction at the position indicated by the point “Pc” on the runway is performed. That is, the actual azimuth is out of the prediction range, and in this example, the current position link is changed to the forward link, and the position correction is performed.

  FIG. 10 is a flowchart illustrating the procedure for acquiring node position data and determining the road alignment of a single road and the procedure for controlling the driving support device using linear data related to the road shape.

  First, in step S1, necessary information shown below is read into the apparatus.

・ Road map data used in the navigation system ・ GPS data ・ Vehicle speed data ・ Direction data (rate gyro detection information)

  In the next step S2, map matching is performed according to a predetermined procedure to determine a road on which the host vehicle is currently traveling, and to estimate the current position of the host vehicle on the traveling road. At the same time, a plurality of node groups including the current node Ni, for example, (Ni-1, Ni, Ni + 1, Ni + 2, Ni + 3, Ni + 4) candidates and a plurality of link groups, for example, Candidates (Li-1, Li, Li + 1, Li + 2, Li + 3) are set, and node position and link vector data are collected.

  In the case of a single road, two adjacent nodes are obtained based on the GPS coordinates, and the place that is the shortest point on the link between the two nodes from the GPS coordinates is set as the vehicle current position after matching. In addition, when the road is not a single road (for example, when there are a plurality of roads in the vicinity of the vehicle such as a hairpin curve or an urban area), the current vehicle position is temporarily estimated on the link with the highest reliability based on the GPS coordinates and the direction data. decide. In addition, a method of constantly or periodically comparing the travel locus obtained from GPS data, vehicle speed, and direction data with the road shape on the road map (because there is a fixed relationship between the steering angle and the turning radius, Steering angle data and travel distance data are used to obtain data indicating the travel locus of the host vehicle, so that the steering operation amount obtained from the road alignment based on the map data and the actually detected steering operation amount. Can be used to determine whether or not the difference between the two is within a predetermined range.), The estimation accuracy of the current position can be further increased.

  In step S3, the section characteristics (that is, the curved section, the inflection section, and the straight section) of the current position link Li are analyzed. For example, when the above form (II) is adopted, the link length | Li | and the outer product | Li-1 × Li |, | Li × Li + 1 | are calculated, and the sign change of the outer product is examined to determine which Li is A primary determination is made as to whether it corresponds to a section.

  In step S4, the section characteristics (that is, the curved section, the inflection section, and the straight section) of the forward link Li + 1 are analyzed. For example, when the above form (II) is adopted, the link length | Li + 1 | and the outer products | Li × Li + 1 | and | Li + 1 × Li + 2 | are calculated, and the sign change of the outer product is examined. Thus, it is primarily determined which section Li + 1 corresponds to.

  In step S5, based on the primary determination results of Li and Li + 1, as described with reference to FIG. 7, the travel sections are discriminated while referring to the combination of both, and a road linear shape equation is created.

  For example, processing is performed according to the following algorithm.

(A1) When both Li and Li + 1 are curved sections, that is, the link lengths | Li | and | Li + 1 | are the reference value (a predetermined multiple of the second speed converted value of the road legal speed, Is described as “Lmin”.) In the following case, the road shape equation is obtained using 3-point circular interpolation, bisectoral circle method, or Newton 4-point forward interpolation method.

(A2) If both Li and Li + 1 are straight sections, that is, the link lengths | Li | and | Li + 1 | are the reference value (a predetermined multiple of the second speed converted value of the road legal speed, In the above case, when the link angle between Li and Li + 1 is “θi” and the reference angle is “θmax”, if “θi ≧ θmax”, It is determined that the bending path has Ni-1 as the bending point, and a smoothing process is performed by applying the contact circle (R) as described above, and the curve starting point and end point are obtained.

  The relationship between the road design speed and the linear radius (curvature radius) on the road design is illustrated in the table below.

(A3) In cases other than the above (A1) and (A2) (hereinafter, the distance from the current position of the vehicle to the forward node Ni + 1 will be referred to as “D”.)
(A3-1a) When “Li ≧ Lmax” and “D ≧ Lmin”, a straight road is assumed.

  (A3-1b) When “Li ≧ Lmax” and “D <Lmin”, any of the following is adopted to define the road shape equation.

(Part 1) Ni + 1 is the bending point, and the running path shape is determined by applying the bending angle θi of the bending road and the tangent radius of Table 3 above, and the curve starting point and end point are obtained. (Part 2) Li Is a straight section, and Ni + 1, Ni + 2, and Ni + 3 are curved sections, and three-point circular interpolation is performed.

  (A3-2a) When Li is an inflection section of “Li <Lmax” and “D ≧ Lmin”, a straight road is assumed.

  (A3-2b) When Li is an inflection section of “Li <Lmax” and “D <Lmin”, one of the following is adopted to define the road shape equation.

(Part 1) Ni + 1 is the bending point, and the running path shape is determined by applying the bending angle θi of the bending road and the tangent radius of Table 3 above, and the curve starting point and end point are obtained. (Part 2) Li Is a straight section, and Ni + 1, Ni + 2, Ni + 3 are curved sections, and three-point circular interpolation is performed. (Part 3) The relocation point Pi-1 outside the nodes Ni and Ni + 1 with respect to the link Li , Pi + 2 is created, and Newton 4-point forward interpolation is performed.

  (A3-3) When Li is a curved road section of “Li <Lmax” and Li + 1 is a straight section or an inflection section, either of the following is adopted to calculate the road shape equation. Define.

(Part 1) Using Ni + 1 as the ending point and the bending point of the road, calculate the bending angle θi of the bending road at Ni + 1, or apply the tangent radius shown in Table 3 above to Ni + 1 and runway shape (Part 2) Create nodes corresponding to the relocation points Pi-1 and Pi + 2 outside the nodes Ni and Ni + 1 with respect to the link Li, respectively. Newton 4-point forward interpolation.

  Note that the above processing is merely an example, and it is needless to say that the present invention can be implemented in various modes.

  In step S6, the map matching is checked for misidentification, and the current position is corrected in step S8. The azimuth range at the link Li is calculated using the road shape equation and compared with the rate gyro (azimuth sensor) data. If there is no problem in the current position including the error of the rate gyro, the process proceeds to step S7. If the difference between the two is large, the process proceeds to step S8, and the position on the track closest to the direction data within the error range of the rate gyro is determined. Set as the current vehicle position. In that case, the direction of the link Li-1 or Li + 1 in the vicinity of the current position link is compared with the direction detected by the rate gyro, and the link with the smaller difference is determined as the current position link. Adopt (that is, move the current node forward or back one). Then, the process returns to step S3. Note that steps S6 and S8 are not required in a configuration in which position misrecognition can be prevented by always comparing the travel trajectory obtained from GPS data, vehicle speed, and direction data with the road shape on the road map.

  In step S7, information necessary for the driving support device, for example, driving operation data (detection data of the steering sensor, etc.) is read. Further, GPS data, vehicle speed data, direction data, and the like are read as necessary. In the next step S9, the following processing is performed.

・ Change the current position of the vehicle according to the change in the travel distance of the vehicle with the passage of time from the map matching ・ Determine the target control amount to the driving support device ・ The above target control amount and the current amount Control the driving support device so that the difference decreases.

  A control target at the current position of the vehicle can be calculated by headlamp light distribution control, vehicle steering, deceleration model, and the like, and driving support can be realized by known control to achieve the control target. For example, in an illumination support system using a light distribution control type headlamp device, the direction of the irradiation light, the irradiation range, the light distribution, etc. can be freely changed, and a predetermined traveling time has elapsed for the own vehicle. It is possible to control the irradiation direction and irradiation range by predicting the vehicle arrival position after or after the mileage, and to control the irradiation direction and the irradiation range, or to control the irradiation direction with the clipping point of the curve as a gazing point . The clipping point is a contact point when a shoulder line or a center line is obtained from the shape line of the running road and a tangent is drawn from the vehicle position with respect to the shoulder line or the center line (the shoulder line or the center line and the tangent line). In addition, the angle of the clipping point means an angle formed between an axis extending in the traveling direction of the vehicle and a tangent drawn from the vehicle position to the shoulder line or the center line, It can be obtained by a known method (tangent circle method) using road shape data (linear data).

  In addition, it is possible to predict a vehicle arrival position after a predetermined travel time or a travel distance with respect to the own vehicle, issue a warning such as approaching a curved road, a warning, and the like, and display various information for driving support.

  In the next step S10, if the elapsed time up to the present time is outside the map matching period, the process returns to step S2, and if it is within the period, the necessity of changing the current node is determined in the next step S11. The process returns to step S3, but if not necessary, the process proceeds to step S12.

  In step S12, it is determined whether or not the range on the road map needs to be changed. If necessary, the process returns to step S1, but if not necessary, the process returns to step S7.

  FIG. 11 shows an example of a system configuration using an ECU (electronic control unit) incorporating a computer. Internal processing of the ECU (for example, the section determination and interpolation processing described above) is performed as software processing using a CPU (central processing unit), memory, hardware such as an input / output port, and a program executed by the CPU. .

Examples of ECUs connected by an in-vehicle LAN (Local Area Network) include the following (numbers in parentheses indicate symbols).
-Operation information acquisition ECU (13)
・ Navigation ECU (14)
-ECU for VSC (15)
-Beam control ECU (16)
-ATC ECU (17).

  The operation information acquisition ECU 13 (for example, a column ECU provided in the steering column) has steering information in addition to operation information such as a headlamp turn-on / off instruction and beam switching instruction, and an operation instruction to a direction indicator. Detection information of the rotation angle of the wheel is input from the steering sensor. These pieces of information are notified to other ECUs through the LAN.

  GPS information or road-to-vehicle communication information, road map information, vehicle speed information, and the like are input to the navigation ECU 14 used for route guidance and the like.

  The VSC ECU 15 receives detection information from a vehicle speed sensor, a yaw rate sensor, or a steering sensor, and performs, for example, travel support control for avoiding a spin or the like when traveling on a curved road.

  The beam control ECU 16 performs irradiation control of the headlamp according to the traveling state of the host vehicle, acquires information necessary for control from other ECUs, the irradiation range, irradiation direction, and light quantity of the headlamp. Control etc.

  The ATC ECU 17 receives vehicle state detection information, vehicle speed information, driving operation information, and the like, and acquires necessary information from other ECUs so as to obtain a gear ratio and driving force suitable for the traveling state of the host vehicle. Then, a transmission (such as a continuously variable transmission) and a drive source (such as an engine and a motor) are controlled.

  Information necessary for beam control of the vehicle headlamp is operation information from the ECU 13 and various information from the ECU 14 ECU 15 and the like. A navigation device, a GPS sensor, and a direction sensor for setting and searching a route and route guidance. A road map information providing device (such as a drive device using an optical recording medium), a vehicle speed sensor, or the like is used.

  Examples of the control configuration include the following forms.

(A) Configuration form in which nodes identified as runway shape change points are extracted so that the irradiation control amount does not change suddenly before and after that (b) When the runway shape changes (curves, bends, intersections, etc.) ), A configuration form in which the irradiation control content is changed according to the approach position of the host vehicle to the shape change point (c) a configuration mode in which the irradiation control content is changed according to the characteristics of the runway section

  First, in (a), the position data of a plurality of nodes positioned before and after the current position of the own vehicle is read using the current position data and road map data of the own vehicle, and the plurality of nodes are connected by interpolation processing to form a running path shape. And the node identified as the change point of the runway shape is grasped. When the host vehicle travels in a range including before and after the node, the control amount related to the irradiation control of the vehicle headlamp is gradually changed.

  It is desirable to gradually change the irradiation state before the change by the control signal sent from the beam control ECU 16 to the driving means (actuator, drive circuit, etc.) of the headlamp, and then shift to the irradiation state after the change. In other words, suddenly changing the irradiation state when the runway shape changes has a large effect on the driver and road users, so the transition period for gradually changing the light distribution, brightness, beam irradiation direction, irradiation range, etc. It is desirable to shift to a new irradiation state with a continuous change (for example, in the case of a device having a function of changing the beam irradiation direction by controlling the optical axis of the lamp, from the irradiation direction before the change) Control may be performed so that the direction is continuously changed to the irradiation direction after the change.)

  In the above (b), for example, a control mode for performing irradiation control of the vehicle headlamp while estimating the road shape around the current vehicle position using the current vehicle position data and road map data, and the current vehicle The control mode for performing the irradiation control of the vehicle headlamp based on the estimation result of the road shape around the position and the detection information of the steering state of the host vehicle is configured to be switched according to the traveling state. Then, the travel distance or travel time to the curved road, the curved road, and the intersection predicted to exist in the traveling direction with the current vehicle position as a reference is calculated and compared with a predetermined reference value. One of the control modes is selected according to the comparison result.

  In (c) above, the shape of the runway ahead of the host vehicle is estimated, and if there are curved roads, bent roads, intersections, etc. on the runway in the direction of travel of the host vehicle, from the entry point of the own vehicle to the exit point Is divided into a plurality of sections, and irradiation control of the vehicle headlamp is performed based on the characteristics of each divided section and the detection information of the vehicle steering state. For example, in the approach section to the curved road, the control amount of the irradiation direction set at the entry point to the curved road is compared with the control quantity of the irradiation direction calculated based on the detection information of the steering state. And if the irradiation direction of a vehicle headlamp is controlled according to the control amount of the larger one of both, the front illumination according to the actual situation will be obtained. Further, in the track section where the steering state in a certain direction is maintained, the irradiation direction of the headlamp is controlled according to the irradiation control amount calculated based on the vehicle steering state detection information.

In addition, regarding the sharing of the above processing in the ECU, for example, the following configuration forms can be cited.
The navigation ECU 14 simply outputs map data, current position data, and the like, and the beam control ECU 16 performs all processes including the prediction of the travel path. Including the start point, end point, inflection point, etc.) and link characteristics (curvature, linearity, flexibility, link interval, etc.), and outputs the calculation results. The received beam control ECU 16 performs irradiation control of the headlamp.

  Next, the ATC will be described. Information necessary for the ATC includes the following information regarding the vehicle state, detection information of the current position of the vehicle from the navigation ECU 14, road map information, and the like. A force control device or the like is used.

-Accelerator operation information-Brake operation information-Vehicle speed detection information-Throttle opening detection information-Input rotation speed detection information-Engine rotation speed detection information-Oxygen concentration detection information-Intake amount detection information

  The flow of the optimum gear ratio calculation and drive control in the ECU 17 is generally as follows.

(S1) Acquire road information of own vehicle traveling path (s2) Collect vehicle information (s3) Calculate recommended vehicle speed from information of (s1) and (s2) (s4) To recommended vehicle speed of (s3) Calculate the deceleration required for the change (s5) Calculate the optimum transmission ratio of the continuously variable transmission based on the deceleration of (s4). (S6) Control the continuously variable transmission and the driving force.

  In the shift control, the following processes (1) to (9) are performed.

(1) Input of vehicle information (vehicle speed, brake operation, input speed, engine speed, throttle opening, accelerator operation, etc.) (2) Input of optimum gear ratio (3) Normal gear ratio determination (4) Optimal speed ratio and Comparison judgment with reference gear ratio (calculated from throttle opening and vehicle speed) (“optimum gear ratio> reference gear ratio” proceeds to (5), but “optimum gear ratio ≦ reference gear ratio” (In this case, after issuing a gear ratio command to the reference gear ratio, the process returns to (1).)
(5) Calculation of current output based on engine speed and throttle opening (6) Calculation of optimum throttle opening (7) Gear ratio command to optimum gear ratio (8) Throttle opening command (9) (1) Return.

  In ATC, the optimum transmission ratio is calculated based on the road shape data, and the operation control of the automatic transmission is performed, and the operation control of the drive source is performed so that the output of the drive source is maintained constant. That is, the gear ratio and the driving force are controlled appropriately for the shape of the vehicle's running path, and thereby it is possible to realize driving support that does not give a sense of incongruity to the driver's accelerator operation or the like. In this case as well, as described above, for example, a change point or the like of the road shape is identified, and control is performed so that the gear ratio or the like does not change suddenly before or after the change. Various configurations such as changing the control content according to the distance to the place and the travel time or changing the control content according to the characteristics of the travel road section are possible.

  In addition, in application to the stability control system and traction control system, it is necessary to accurately grasp the road alignment of the traveling path of the host vehicle. By applying the present invention, it is possible to travel on a curved road. It is effective in enhancing stability (effective in preventing accidents by accurate driving support).

  As described above, in the present invention, the current position data of the vehicle is acquired and the position data of a plurality of nodes positioned before and after the current position of the vehicle is read using the road map data. Then, the characteristics of each link formed by connecting adjacent nodes, that is, the shape characteristics including the linearity and the curvature characteristics of the runway are determined. The road alignment is estimated by connecting the plurality of nodes according to an appropriate interpolation process used according to the determination result, and the driving support device is controlled according to the estimation result of the road shape predicted ahead of the host vehicle. Accordingly, it is possible to prevent the adverse effects of applying improper interpolation to the inflection section or the like without considering the characteristics of the runway section.

  According to the configuration described above, the following advantages can be obtained.

  ・ Accurately grasp the shape characteristics related to the road alignment of the road required for driving support control of the vehicle and obtain the shape data in a short time (as a result, the control accuracy is improved and the driver feels strange No control can be realized.).

  -Since there is little burden on the calculation processing to the CPU or the like, it is not necessary to increase the processing capacity of the ECU or the like more than necessary, which is advantageous in terms of cost.

It is a figure which shows the structural example to which this invention is applied. It is explanatory drawing about the definition of a node and a link. It is explanatory drawing of a three-point circular interpolation method. It is explanatory drawing of a bisection circle method. It is explanatory drawing of Newton's 4-point forward interpolation method. It is explanatory drawing of a smoothing process. It is a figure which shows the example of a runway shape definition by a table format. It is the figure which illustrated the runway shape. It is explanatory drawing regarding present position correction. It is a flowchart figure which shows the example of control. It is a figure which shows the system configuration example using ECU. It is a figure which shows a S-shaped curved path as an example of a road alignment shape.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Vehicle travel assistance system, 2 ... Own vehicle present position detection means, 4 ... Node data position extraction means, 5 ... Link characteristic discrimination means, 6 ... Road alignment generation means, 7 ... Control means, 9 ... Vehicle current position correction means

Claims (7)

A vehicle travel support system for controlling a travel support device of a vehicle according to an estimation result of a road shape predicted in front of the host vehicle,
Own vehicle current position detecting means for acquiring current position data of the vehicle using satellite communication or road-to-vehicle communication;
Node data position extraction means for reading position data of a plurality of nodes located before and after the current position of the vehicle using the current position data and road map data of the vehicle acquired by the current position detection means;
Link characteristic determining means for determining the shape characteristic of each link formed by connecting adjacent nodes among the plurality of nodes and the type of the runway section;
Road linear generation means for generating road shape data modeled by connecting the plurality of nodes by interpolation processing;
Control means for sending a control output according to the estimation result of the road shape to the driving support device;
The type of the road segment determined by the link characteristic determination unit is one of a plurality of categorized road segments, and the link characteristic determination unit determines the type of the road segment and determines the determination result as the road. Sent to the linear generator,
The road alignment generation means performs the interpolation processing according to the determination result determined by the link characteristic determination means. The vehicle travel support system according to claim 1,
The vehicle travel support system, wherein the travel support device is an irradiation control device for a vehicle headlamp. The vehicle travel support system according to claim 1,
The vehicle travel support system characterized in that the type of the road section discriminated by the link characteristic discriminating means is either an inflection section including an inflection point of a road shape, a curved section or a straight section. The vehicle travel support system according to claim 1,
The type of the road section discriminated by the link characteristic discriminating means compares the link length between adjacent nodes with a reference distance value determined in advance based on the second speed converted value of the legal running speed, and the link length is When the link distance is equal to or greater than the reference distance value, the track section related to the link is determined as a straight section, and when the link length is less than the reference distance value, the track section related to the link is the inflection section or the curve section. A vehicle travel support system characterized by being discriminated. The vehicle travel support system according to claim 1,
The type of the road section discriminated by the link characteristic discriminating means is discriminated by calculating the sign change of the outer product of the link vectors obtained as the difference between the link length of the link between adjacent nodes and the position vector of the node. A vehicle travel support system characterized by the above. The vehicle travel support system according to claim 1,
The type of the runway section discriminated by the link characteristic discriminating means is discriminated by obtaining the radius of curvature obtained by performing circular interpolation on the three nodes constituting two adjacent links and the sign change of the radius of curvature. A vehicle driving support system characterized by being provided. A vehicle travel support system that controls a travel support device of a vehicle according to an estimation result of a road shape predicted in front of the host vehicle,
Own vehicle current position detecting means for acquiring current position data of the vehicle using satellite communication or road-to-vehicle communication;
Node data position extraction means for reading position data of a plurality of nodes located before and after the current position of the vehicle using the current position data and road map data of the vehicle acquired by the current position detection means ;
Link characteristic determining means for determining the shape characteristic of each link formed by connecting adjacent nodes among the plurality of nodes and the type of the runway section;
Road linear generation means for generating road shape data modeled by connecting the plurality of nodes by interpolation processing;
Control means for sending a control output according to the estimation result of the road shape to the driving support device;
Vehicle current position correction means for correcting the current vehicle position data is provided;
Based on the road alignment shape equation estimated from the road shape data generated by the road alignment generation means, the direction change prediction range in the current link is calculated, and the direction data of the host vehicle is out of the prediction range. In this case, it is determined that the current position of the host vehicle does not exist on the current link, the target is changed to another link located around the current link, and the correction process of the current position of the host vehicle is continued. When the azimuth data is included in the predicted range, a position that is the same as the azimuth data on the current link or closest to the data on the current link is estimated as the current vehicle position. A vehicle driving support system.

Images