JP6542153B2 - Route generation apparatus, route generation system, route generation method and route generation program - Google Patents

Description

  The present invention relates to a route generation device, a route generation system, a route generation method and a route generation program for generating a route based on an instruction from an occupant.

  The technology of automatic driving is developing, and in-lane automatic driving may be realized in the near future (Non-Patent Document 1). However, in the case of in-lane automatic driving, an instruction from the occupant is required when changing the lane or turning left or right at the intersection.

  For example, if you make a left turn at the previous intersection while traveling on the rightmost lane of a road with multiple lanes, the occupant repeats the left lane change instruction until the leftmost lane is reached, and then instructs the left turn. It has to be bothersome.

https://www.teslamotors.com/jp/blog/your-autopilot-has-arrived

  The present invention has been made in view of such problems, and an object of the present invention is to provide a route generation device, a route generation system, a route generation method, and a route generation capable of generating an appropriate route by a simple instruction from an occupant. It is to provide a program.

According to one aspect of the present invention, a lane change receiving unit that receives a lane change of a vehicle from an occupant, a vehicle state acquisition unit that acquires a state of the vehicle, a map information acquisition unit that acquires map information around the vehicle, There is provided a route generation device including: a route estimation unit configured to estimate a route after a lane change based on the state of the vehicle and the map information when a lane change is received.
Since the route can be appropriately generated by considering the condition of the vehicle, the instruction from the occupant can be simplified.

  When there is a turning point in the traveling direction of the vehicle, the route estimation unit may estimate the path at the turning point.

  The state of the vehicle may include at least one of the distance to the junction, the lane in which the vehicle is currently located, and the speed of the vehicle.

  The lane change receiving unit may receive the lane change during automatic driving of the vehicle.

  The route estimation unit estimates the route after the lane change in consideration of at least one of a road traffic condition around the vehicle, a road characteristic around the vehicle, a travel history of the occupant, and a destination. Good.

  The lane change receiving unit may receive a lane change instruction of a vehicle from the occupant via an input interface.

  The route estimation unit may estimate the route after the lane change based on the relationship between the state of the vehicle and the probability that the route after the lane change is selected.

Preferably, the route generation device includes a probability update unit that updates the probability based on the driver's driving history.
In this way, it is possible to estimate a route that suits the occupant.

  According to another aspect of the present invention, there is provided a route generation system comprising: an interface for receiving an instruction of a vehicle traveling direction from an occupant, the route generation device, and an output interface for presenting a route estimated by the route estimation unit. Is provided.

  According to another aspect of the present invention, there is a step of receiving a lane change of a vehicle from an occupant, a step of acquiring a state of the vehicle, a step of acquiring map information around the vehicle, and the lane change accepted. And E. estimating a path after a lane change based on the state of the vehicle and the map information.

  According to another aspect of the present invention, the computer receives a lane change of a vehicle from an occupant, a step of acquiring a state of the vehicle, a step of acquiring map information around the vehicle, and the lane change Providing a route generation program that executes a step of estimating a route after a lane change based on the state of the vehicle and the map information.

  An appropriate route can be generated by a simple instruction from the occupant.

The figure explaining operation of one embodiment. 1 is a block diagram showing a schematic configuration of a route generation system according to a first embodiment. The figure which shows typically probability P1 (s) which performs instruction | indication of a left change in the position s, when vehicle 100 in the right lane L2 changes lanes to left lane L1, and turns left at intersection C1 after that. When the vehicle 100 in the right lane L2 turns left, probability P1 (s) of giving a left change instruction at position s and probability P2 (s) of giving a left change instruction at position s when changing the left lane as well The figure shown typically. The figure which shows typically probability P (s, v) which performs instruction | indication of left change by the position s and the vehicle speed v, when the vehicle 100 in the right lane L2 turns left. BB sectional drawing of FIG. 5A. CC sectional drawing of FIG. 5A. The figure which shows typically another example of FIG. The block diagram which shows schematic structure of the path | route production | generation system which concerns on 2nd Embodiment. The figure explaining update of probability.

  Hereinafter, embodiments according to the present invention will be specifically described with reference to the drawings.

First Embodiment
FIG. 1 is a diagram for explaining the operation of one embodiment. In the present embodiment, it is assumed that there is an intersection C1 of a left turn in the traveling direction of the vehicle 100 on a one-way two-lane (L1, L2) road. It is assumed that an instruction to move in the left direction at a position where the occupant of the vehicle 100 in the right lane L2 is present is issued to the route generation device according to the present embodiment. In this embodiment, in such a case, the route generation device 2 takes the driver's intention, changes lanes, and then turns left at the intersection C1 (route R1) or selects only lane change (route R2). It is As a result, even if the driver changes lanes and turns left at intersection C1, the driver only needs one instruction.

  FIG. 2 is a block diagram showing a schematic configuration of a route generation system according to the first embodiment. The route generation system includes an input interface 1, a route generation device 2, and an output interface 3. The route generation system is provided in the vehicle 100 and can be used, for example, for automatic driving.

  The input interface 1 is used by an occupant to instruct the route generation device 2. The input interface 1 is preferably a winker, a lever, a joystick, a touch pad, a microphone, etc. that can be instructed by a simple operation.

  The route generation device 2 includes a lane change reception unit 21, a vehicle state acquisition unit 22, a map information database 23, a map information acquisition unit 24, and a route estimation unit 25. All or a part of each unit except the map information database 23 may be implemented by hardware, or may be realized by a predetermined route generation program being executed by a processor.

  The lane change receiving unit 21 receives an instruction from the occupant via the input interface 1. This instruction may be issued during automatic driving of the vehicle 100, more specifically, during in-lane automatic driving.

  The lane change receiving unit 21 of the present embodiment receives an instruction to change the left or right from the occupant as an instruction to change the lane. Even if the passenger simply changes lanes to the left lane or turns left after changing to the left lane, he / she may issue a left change instruction once.

  The vehicle state acquisition unit 22 acquires the state of the vehicle 100. The state of the vehicle 100 may include the position of the vehicle 100 acquired by a GPS receiving device mounted on the vehicle 100. Here, the distance from the vehicle 100 to each intersection can be calculated by combining the position of the vehicle 100 and map information to be described later, so the position of the vehicle 100 and the distance from the vehicle 100 to each intersection are considered equivalently It can also be done. The state of the vehicle 100 may be the vehicle speed, and can be acquired from, for example, a vehicle speed sensor mounted on the vehicle 100. Furthermore, the state of the vehicle 100 may include in which lane the vehicle 100 is currently located. For example, the vehicle state acquisition unit 22 can identify the current lane by using image recognition using an on-vehicle camera, position information of the vehicle 100 acquired by the GPS receiver, map information, and the like.

  The map information database 23 stores map information. The map information includes the shape of the road and the number of lanes, and may further include real-time road traffic conditions. Although FIG. 2 shows an example in which the map information database 23 is built in the route generation device 2, the map information database 23 may be a storage medium separate from the route generation device 2, or the route generation device 2 may be connected via a network.

  The map information acquisition unit 24 acquires map information around the vehicle 100. More specifically, the map information acquiring unit 24 grasps the own position of the vehicle 100 based on the information from the GPS receiving device, and acquires map information including the road around it, particularly, the road on the traveling direction side.

  The route estimation unit 25 estimates a route of the vehicle 100 intended by the occupant based on the state of the vehicle 100 and the map information, using the reception of the instruction to change the left and right from the occupant as a trigger. The estimation method will be described later. In the present embodiment, the route estimation unit 25 estimates whether to go straight or turn left at the intersection in the traveling direction after the lane change.

  The output interface 3 presents the route estimated by the route estimation unit 25. The output interface 3 is typically a display provided on a dashboard or the like of the vehicle 100, and displays, for example, an estimated route with an arrow superimposed on map information.

  For the route presented to the output interface 3, the occupant may be able to indicate whether the route is correct via the input interface 1. When the lane change acceptance unit 21 receives an instruction indicating that the route is correct, the route estimation unit 25 may approve the route. If the lane change accepting unit 21 accepts an instruction that the route is not correct, the route estimating unit 25 may estimate an alternative route.

  Thereby, the automatic driving of the vehicle 100 continues along the generated route, and more specifically, the automatic driving is continued even after a lane change or after a lane change after turning to the left or right.

The process of the route estimation unit 25 will be described with reference to FIG. In FIG. 1, when the occupant of the vehicle 100 in the right lane L2 issues an instruction to change the left, the distance from the vehicle 100 to the intersection C1 is s1, and the vehicle speed at that time is v. For example, if the distance s1 is sufficiently small or sufficiently large, it is considered that the passenger intends to go straight at the intersection C1 after changing the lane to the left lane L1. Lane change (route R2). On the other hand, if the distance s1 is within the predetermined range according to the vehicle speed v, it is considered that the occupant intends to turn left at the intersection C1, so the route estimation unit 25 turns left at the intersection C1 after changing lanes to the lane L1. It is estimated that (path R1).
That is, the route estimation unit 25 can estimate the route intended by the occupant from the vehicle speed v at the time of giving the instruction or the distance necessary for sufficiently decelerating at the intersection.

  It will be described more specifically. First, assuming that the state of the vehicle 100 acquired by the vehicle state acquisition unit 22 is only the distance s1 to the intersection (that is, the vehicle speed v is not acquired), the relationship with the selected intersection will be described.

FIG. 3 shows that the vehicle 100 in the right lane L2 changes lanes to the left lane L1 and then turns left at the intersection C1 (in the case where the route R1 in FIG. 1 is intended, hereinafter simply referred to as left turn) It is a figure which shows typically the probability P1 (s) which performs instruction | indication of a change in the position s. The horizontal axis is the position of the vehicle 100 at the time of giving a left change instruction, and the distance between each position and the intersection C1 corresponds to the distance from the vehicle 100 to the intersection C1. The vertical axis indicates the probability P1 (s) that the left change instruction is given at each position. That is, FIG. 3 shows at which position the left change instruction is likely to be given when it is desired to turn left at the intersection C1. For example, when it is intended to turn left at intersection C1, it is indicated that the probability that a left change instruction is given at position sa (distance sa1 with intersection C1) is P1 (sa). In other words, when a left change instruction is given at the position sa, the probability that the occupant intends to turn left at the intersection C1 is P1.
Such probability distribution is previously set in the route estimation unit 25 for each intersection.

  Similarly, when the vehicle 100 in the right lane L2 changes lanes to the left lane L1 and then goes straight ahead at the intersection C1 (in the case where the route R2 in FIG. 1 is intended, hereinafter, simply referred to as left lane change) , The probability that the left direction is given at the position s, in other words, when the left change instruction is given at the position s, the probability that the occupant intends to change the left lane It is set.

  FIG. 4 shows a probability P1 (s) of giving a left change instruction at position s when the vehicle 100 in the right lane L2 turns left, and a probability P2 of giving a left change instruction at the position s when changing the left lane as well. It is a figure which shows (s) typically. The probability P1 (s) when turning left is indicated by a solid line, and the probability P2 (s) when changing the left lane is indicated by a broken line. Based on these probability distributions, when an instruction to change the left is received from the occupant, the route estimation unit 25 selects either the left turn or the change to the left lane as follows.

  For example, it is assumed that a left change instruction is received from the occupant at the position sa (the distance to the intersection C1 is sa1). The probability that the left change instruction performed at this position sa intends to turn left at the intersection C1 is P1 (sa), and the probability that the left lane change is intended is P2 (sa). Then, since P1 (sa) <P2 (sa) from FIG. 4, the route estimation unit 25 selects the left lane change (route R2).

  As another example, it is assumed that a left change instruction is received from the occupant at the position sb (the distance to the intersection C1 is sb1). Since P1 (sb)> P2 (sb) from FIG. 4, the route estimation unit 25 selects the left turn (route R1) at the intersection C1.

  In this manner, the route estimation unit 25 determines the position (in other words, the distance to each intersection) s, the probability that a left (or right, the same applies hereinafter) change instruction at each position s intends to turn left, and the left lane change. The relationship with the intended probability is stored in advance. Then, the path estimation unit 25 estimates that the path with the largest probability is the path intended by the passenger at the position s at the time of receiving the direction of travel instruction. For example, Bayesian estimation can be applied as a specific estimation method. In addition, the same probability P (s) may be set for intersections having similar shapes and the like.

Subsequently, it will be described that the route is estimated by further considering the vehicle speed v.
FIG. 5A is a diagram schematically showing the probability P (s, v) of instructing the left change at the position s and the vehicle speed v when the vehicle 100 in the right lane L 2 turns left. The horizontal axis is the position s of the vehicle 100 at the time of instructing the left change, which is the same as the horizontal axis in FIG. 3. Further, the vertical axis is the vehicle speed v at the time of instructing the left change. Then, the probability P (s, v) that the left change instruction is given at each position s and each vehicle speed v is shown in the direction perpendicular to the drawing sheet.

  The probability P (s, v1) in FIG. 5B and the probability P (s, v) in FIG. 5A cut by line B-B (vehicle speed v1) and line C-C (vehicle speed v2 (<v1)). Each corresponds to the probability P (s, v2) of 5C. Such probability distribution is set in advance in the route estimation unit 25 for each intersection and each lane.

  As described above, the route estimation unit 25 determines the traveling direction change instruction at the position (in other words, the distance to each intersection) s, the vehicle speed v, and each position s and each vehicle speed v P (s, v) Are stored in advance for each route (that is, for each intersection and lane). Then, the route estimation unit 25 estimates and selects a route with the largest probability P (s, v) as the route intended by the occupant at the position s and vehicle speed v at the time of receiving the direction of travel instruction. May be

Next, an example of the setting method of the probability distribution shown in FIG. 3 and FIG. 5A will be described.
As a first example, the probability distribution may be set in consideration of the road traffic condition around the vehicle 100. In this case, the route estimation unit 25 selects candidate intersections taking into consideration the road traffic condition. . As a specific example, the route estimation unit 25 can recognize and consider the traveling direction of the vehicle 100 or the traffic volume at the intersection, and the number of vehicles waiting for turning at the intersection. For example, in FIG. 1, when the left turn destination of the intersection C1 is not crowded but the straight ahead destination is crowded, the probability of turning left at the intersection C1 is made high overall, and instead of FIG. It may be a probability distribution. In this case, the probability distribution is not fixed and may vary depending on the road traffic situation.
In the case of the probability distribution as shown in FIG. 6, when a left change instruction is given at the position sa, P1 (sa)> P2 (sa), so the route estimation unit 25 does not change the left lane (route R2). , Left turn (route R1) to choose.

  As a second example, a probability distribution may be set in consideration of general preferences, specifically, road characteristics around the vehicle 100. In this case, the route estimation unit 25 may be a candidate in consideration of the road characteristics. Would be to select the intersection. As a specific example, the route estimation unit 25 can consider the number and width of lanes of the road included in the map information, the amount of traffic, and whether or not it is a toll road. For example, in FIG. 1, when the road to which the intersection C1 turns left is wide but the straight road is narrow, the probability distribution as shown in FIG. 6 may be used instead of FIG.

  As a third example, the probability distribution may be set in consideration of individual preference of the occupant, specifically, the traveling history of the occupant of the vehicle 100. In this case, the route estimation unit 25 considers the traveling history of the occupant. And select a candidate intersection. As a specific example, the route estimation unit 25 can consider that the road is a road through which the occupant has traveled well in the past, or a feature of a favorite road (such as passing through a wide road). For example, in FIG. 1, the occupant often turns left at the intersection C1, but if there is not much going straight at the intersection C1, the probability distribution as shown in FIG. 6 may be used instead of FIG. In this case, the probability distribution may differ depending on the occupant or depending on the vehicle.

As a fourth example, a probability distribution may be set in consideration of a destination, and in this case, the route estimation unit 25 selects candidate intersections in consideration of the traveling history of the occupant. The destination can be, for example, one set in car navigation. Then, the probability of an intersection having an excellent evaluation value (e.g., a required time until arrival) regarding the route to the destination is set high. For example, in FIG. 1, when there is a destination on the road on the left of the intersection C1 after turning left, the probability distribution as shown in FIG. 6 may be used instead of FIG.
The first to fourth examples described above may be combined, or other factors may be considered.

  As described above, in the first embodiment, when the occupant gives an instruction to change the left or right, the occupant intends to change only the lane or change the lane after turning based on the state of the vehicle 100. Estimate the route. Therefore, the route can be appropriately generated, and in the in-lane automatic driving, the automatic driving can be continued even after the lane change and the right turn.

In addition, this embodiment is applicable not only to an intersection but also when arbitrary junctions or two or more intersections exist in the advancing direction. The present invention is also applicable to a road having three or more lanes.
Further, the reason that the route estimation unit 25 performs the route estimation may not necessarily be a lane change instruction from the occupant via the input interface 1, but the occupant may have changed the lane by his own driving. It is also good. That is, the lane change receiving unit 21 may receive the lane change performed by the occupant, and the route estimation unit 25 may estimate the route after the lane change according to the received lane change. This is particularly effective when the vehicle does not automatically change lanes.

Second Embodiment
The second embodiment to be described next is to update the probability distribution shown in FIG.
FIG. 7 is a block diagram showing a schematic configuration of a route generation system according to the second embodiment. As a difference from FIG. 2, the path generation device 2 ′ further includes a probability update unit 26. The probability update unit 26 acquires the instruction by the occupant and the history of the actual driving, and updates the probability based on the history.

  Some crew members give instructions early, and some crew members give instructions just at the intersection. Therefore, by acquiring the history for each occupant and generating the probability for each occupant, it is possible to generate a route that matches the habit of the occupant.

  FIG. 8 is a diagram for explaining the update of the probability. In the same figure (a), when turning left at the intersection C1, the probability P1 (s) of giving a left change instruction at the position s is predetermined, and the left change instruction is most likely to be issued at the position s0. . This probability P1 (s) is an initial setting that does not depend on the occupant.

  According to the driving history of a certain occupant A, it is assumed that the left change instruction is often given at the position sa0 closer to the intersection C1 than the position s0. The position sa0 can be acquired by the vehicle state acquisition unit 22. In this case, the probability updating unit 26 generates the probability P1a (s) by increasing the probability of the position sa0 in the probability P1 (s) (FIG. 8 (b)). If the instruction to change the left is often issued at the position sa0 thereafter, the probability updating unit 26 further increases the probability of the position sa0 at the probability P1a (s) (FIG. 8 (c)). Thus, the probability P1a (s) for the occupant A is updated.

On the other hand, according to the driving history of a certain occupant B, it is assumed that the left change instruction is often given at the position sb0 which is separated from the intersection C1 from the position s0. In this case, the probability updating unit 26 increases the probability of the position sb0 in the probability P1 (s) to generate the probability P1b (s) (FIG. 8 (d)). If the instruction to change the left is often issued at the position sb0 thereafter, the probability update unit 26 further increases the probability of the position sb0 at the probability P1b (s) (FIG. 8 (e)). Thus, the probability P1b (s) for the occupant B is updated.
Then, the route estimation unit 25 estimates a candidate route using the updated probability.

  Thus, in the present embodiment, the probability distribution is updated based on the history of the occupant. Therefore, it is possible to properly estimate the route suitable for the occupant.

  As shown in FIG. 5A, when the probability P (s, v) of giving the left change instruction at the position s and the vehicle speed v is determined when turning left at the intersection, the left change instruction as the driving history. The vehicle state acquisition unit 22 may acquire the vehicle speed v as well as the position s with respect to the intersection at the time of performing, and the probability update unit 26 may update the probability P (s, v).

  The embodiments described above are described for the purpose of enabling one skilled in the art to which the present invention belongs to to practice the present invention. Various modifications of the above-described embodiment can naturally be made by those skilled in the art, and the technical idea of the present invention can also be applied to other embodiments. Therefore, the present invention should not be limited to the described embodiments, but should be the broadest scope in accordance with the technical idea defined by the claims.

Reference Signs List 1 input interface 2 route generation device 21 lane change reception unit 22 vehicle state acquisition unit 23 map information database 24 map information acquisition unit 25 route estimation unit 3 output interface

Claims (11)

A lane change reception unit that receives a lane change of a vehicle from an occupant;
A vehicle state acquisition unit that acquires the state of the vehicle;
A map information acquisition unit that acquires map information around the vehicle;
A route generation device comprising: a route estimation unit configured to estimate a route after lane change based on the state of the vehicle and the map information when the lane change is accepted.   The route generation device according to claim 1, wherein the route estimation unit estimates a route at a branch point when there is a branch point in the traveling direction of the vehicle.   The route generation device according to claim 2, wherein the state of the vehicle includes at least one of a distance to the branch point, a lane in which the vehicle is currently, and a speed of the vehicle.   The route generation device according to any one of claims 1 to 3, wherein the lane change receiving unit receives the lane change during automatic driving of the vehicle.   The route estimation unit estimates a route after the lane change in consideration of at least one of a road traffic condition around a vehicle, a road characteristic around the vehicle, a travel history of an occupant, and a destination. The path generation device according to any one of Items 1 to 4.   The route generation device according to any one of claims 1 to 5, wherein the lane change receiving unit receives a lane change instruction of a vehicle from the occupant via an input interface.   The route estimation unit according to any one of claims 1 to 6, wherein the route after the lane change is estimated based on the relationship between the state of the vehicle and the probability that the route after the lane change is selected. Route generator.   The route generation device according to claim 7, further comprising: a probability update unit configured to update the probability based on a driver's driving history. An interface where a lane change instruction is input from an occupant;
A route generation device according to any one of claims 1 to 8.
An output interface for presenting the route estimated by the route estimation unit. The lane change acceptance unit accepts a lane change of a vehicle from an occupant;
The vehicle state acquisition unit acquires the state of the vehicle;
A map information acquisition unit acquiring map information around the vehicle;
A route estimation unit that estimates a route after the lane change based on the state of the vehicle and the map information when the lane change is accepted. On the computer
Accepting from the occupant a lane change of the vehicle;
Obtaining the state of the vehicle;
Obtaining map information around the vehicle;
A route generation program for executing a step of estimating a route after a lane change based on the state of the vehicle and the map information when the lane change is accepted.

Images