GB2632923A - Method for detecting obstacles onboard a machine adapted to move on at least one predetermined path - Google Patents

Abstract

An obstacle detection device 14 onboard a machine (e.g. train 1) stores geographical coordinates of a global volume 31 occupied by the train when circulating along a predetermined path. It determines the train’s 3D position, e.g. using a location block (12,Fig.1) comprising a satellite positioning receiver. An on-board sensor (e.g. LIDAR 13,Fig.1) receives echoes of emitted waves from an object E1,E2 and uses them to calculate its direction and distance relative to the train. Using these data, the device determines whether the detected object (e.g. person E2) is located inside the global volume and, if so, triggers an emergency action (e.g. braking). The triggering may additionally depend on an estimated object size or verification of sensor performance. The global volume may be a function of multiple rectangles (Fig.6-7) defining a machine clearance plate 30. Determining whether the object lies within it may account for a position and attitude of the senor (Figs.8-10) or for alternative paths ahead (Fig.5).

Description

DESCRIPTION

Title: Method for detecting obstacles onboard a machine adapted to move on at least one predetermined path

Technical Field:

[0001] The invention lies in the field of detecting obstacles onboard a machine adapted to move on at least one predetermined path, for example a train.

Background art:

[0002] In autonomous driving system or driverless system in railway applications, in case of failure of the signaling system, a driver has to get onboard the train to drive the train manually with visual inspection to a suitable location, e.g. the next station, to unload the passengers. This causes a significant delay in operation, as well as sacrifice of safety level.

[0003] Known existing solutions for obstacle detection use machine learning or deep learning algorithms, which are not safety certifiable or need cameras, which can be effected by the lighting conditions.

[0004] There is thus a need of an obstacle detection solution that alleviates these drawbacks.

Summary of the invention:

To this end, according to a first aspect, the present invention describes a method for detecting obstacles implemented by an electronic obstacle detection device onboard a machine adapted to move on at least one predetermined path; said obstacle detection device including: a database storing definition data, including 3D geographical coordinates, of the global volume occupied by the machine when circulating along the predetermined path; a tele-detection block adapted for emitting waves towards the path in front of the machine, for receiving echoes of emitted waves from at least one object and to calculate from said waves and echoes, direction and distance, relative to the machine, of said object; said method comprising the steps of: a/ determining the current 3D position of the machine; b/ using the tele-detection block, determining direction and distance relative to the machine of at least one currently detected object; c/ as a function of the determined current 3D position, of the determined current direction and distance of the at least one detected object, determining, based upon said stored 3D geographical coordinates of the volume, if the at least one detected object is located inside said volume; d/ in case the at least one detected object is determined located inside said volume, triggering an emergency action.

[0005] This invention enables to detect if there is any obstacle on the path, without having to identify the type of obstacle.

[0006] In embodiments, such a method will also comprise at least one of the following characteristics: [0007] -in case at least one detected object is determined located inside said volume, an emergency action is triggered only after depending on the result of at least one additional step among: estimating a size of the detected object based upon the received echoes, and verifying said estimated size is greater than a threshold size; verifying the performance of the tele-detection block based upon a number of points n with respect to the range r of the tele-detection and threshold size s of objects; [0008] -in case at least one detected object is determined located inside said volume, the following steps are achieved before triggering any emergency action: -i/ the object point inside the volume that is nearest of the machine within the considered clearance envelop is firstly selected, and named point1; -fi/ the number m of the points inside the volume within a predefined distance s of the selected nearest point is determined; -iii/ if m+1< n.p: point1 is not regarded as pertaining to a relevant obstacle: point1 is discarded, no emergency action is triggered, and the next nearest point is then considered by the algorithm iterating from step i; n being the predefined number of theoretically reflected points and p being the percentage of the points reflected and perceived with given integrity level; otherwise if m+1 n.p, point1 is regarded as belonging to a relevant obstacle and emergency action is triggered; [0009] -in case the path upon which the machine moves is further going to separate into at least two alternative paths, the database storing definition data, including 3D geographical coordinates, of the global volume occupied by the machine when circulating along each of the alternative path: -the situation of alternative paths is detected based upon of the determined current 3D position and the 3D geographical coordinates of the two global volume occupied by the machine when circulating along the at least two alternative paths; and -the steps c and d are then performed regarding the at least two alternatives paths; [0010] -wherein the machine is a train and the predetermined path is a track.

[0011] According to another aspect, the invention describes a computer program adapted to be stored in the memory of an obstacle detection device further comprising a microcomputer, said computer program including instructions which, when executed on the microcomputer, implement the steps of a method according to the precedent aspect of the invention.

[0012] According to another aspect, the invention describes an obstacle detection device adapted to be onboard a machine adapted to move on at least one predetermined path; said obstacle detection device including: a database storing definition data, including 3D geographical coordinates, of the global volume occupied by the machine when circulating along the predetermined path; a tele-detection block adapted for emitting waves towards the path in front of the machine, for receiving echoes of emitted waves from at least one object and to calculate from said waves and echoes, direction and distance, relative to the machine, of said object, said method being adapted to implement the following operations of: a/ determining the current 3D position of the machine; b/ using the tele-detection block, determining direction and distance relative to the machine of at least one currently detected object; c/ as a function of the determined current 3D position, of the determined current direction and distance of the at least one detected object, determining, based upon said stored 3D geographical coordinates of the volume, if the at least one detected object is located inside said volume; d/ in case the at least one detected object is determined located inside said volume, triggering an emergency action.

[0013] In embodiments, such a device will also comprise at least one of the following characteristics: [0014] -the obstacle detection device is adapted, in case at least one detected object is determined located inside said volume, to trigger an emergency action depending on the result of at least one additional operation implemented by the device among: estimating a size of the detected object based upon the received echoes, and verifying said estimated size is greater than a threshold size, verifying the performance of the tele-detection block based upon a number of points n with respect to the range r of the tele-detection and threshold size s of objects; [0015] -the obstacle detection device is adapted, in case at least one detected object is determined located inside said volume, to achieve the following operations before triggering any emergency action: -i/ the object point inside the volume that is nearest of the machine within the considered clearance envelop is firstly selected, and named pointl; -ii/ the number m of the points inside the volume within a predefined distance s of the selected nearest point is determined; -iii/ if m+1< n.p: pointl is not regarded as pertaining to a relevant obstacle: pointl is discarded, no emergency action is triggered, and the next nearest point is then considered by the algorithm iterating from operation i; n being the predefined number of theoretically reflected points and p being the percentage of the points reflected and perceived with given integrity level; otherwise if m+1 n.p, pointl is regarded as belonging to a relevant obstacle and emergency action is triggered; [0016] -the database stores definition data, including 3D geographical coordinates, of the global volume occupied by the machine when circulating along each of the alternative path, and the obstacle detection device is adapted, in case the path upon which the machine moves is further going to separate into at least two alternative paths, to detect the situation of alternative paths based upon of the determined current 3D position and the 3D geographical coordinates of the two global volume occupied by the machine when circulating along the at least two alternative paths; and to perform then the operations c and d regarding the at least two alternatives paths.

Brief description of the drawings:

[0017] The invention will be better understood and other characteristics, details and advantages will appear better on reading the following description, given without limitation, and thanks to the appended figures, given by way of example.

[0018] [Fig. 1] Figure 1 is a schematic view of a processing device in an embodiment of the invention; [0019] [Fig. 2] Figure 2 represents steps of an obstacle detection method in an embodiment of the invention; [0020] [Fig. 3] Figure 3 is an illustration of a situation of a train implementing a processing device according an embodiment of the invention circulating on a track; [0021] [Fig. 4] Figure 4 illustrates an embodiment wherein two LI DARs with different fields of view are used; [0022] [Fig. 5] Figure 5 is a top view illustration of a switch situation in an embodiment of the invention; [0023] [Fig. 6] Figure 6 shows a view in a plan P of a train clearance plate; [0024] [Fig. 7] Figure 7 illustrates the construction of the train envelope in an embodiment of the invention; [0025] [Fig. 8] Figure 8 illustrates pitch; [0026] [Fig. 9] Figure 9 illustrates roll; [0027] [Fig. 10] Figure 10 illustre yaw; [0028] [Fig. 11] Figure 11 illustrates the considered segment in order to determine train attitude; [0029] [Fig. 12] Figure 12 illustrates the considered referential in order to calculate the sensor lever arms.

[0030] Identical references can be used in different figures when they designate identical or comparable elements.

Detailed description:

[0031] The description of the invention is performed herebelow referring to an embodiment with a train 1. The train 1 is adapted to circulate on a railway track network comprising a plurality of tracks. Figures 3 and 5 represent schematic views of the train 1 circulating over sections of the railway track network. Train 1 is manually or autonomously driven.

[0032] The train 1 comprises for example a locomotive facing the further portion of the track to be taken by the train and pulling some wagons of the train.

[0033] Figure 1 shows an electronic processing device 10 aboard the train 1. The processing device 10 includes a database 11, a location block 12, a LIDAR block 13 and an obstacle detection block 14.

[0034] The location block 12 is adapted to determine the position of the train 1 each time period T1 (for example T1 is in the range from 100ms to 1s depending on the obstacle detection block 14 and necessity). The location block 12 for example includes a satellite receiver adapted to determine the satellite receiver position based upon positioning signals including known code from satellites, and/or includes a telemeter and/or an inertial unit. The location block 12 for example provides the 3D absolute coordinates.

[0035] The LIDAR (« Laser Imaging Detection and Ranging ») block 13 is installed on the front of the locomotive, facing the portion of the track the train 1 is going towards, as shown in figure 3.

[0036] As known, the LIDAR block 13 includes at least one laser source adapted to emit a laser pulses and includes sensors. Emitted laser pulses, when meeting objects, are reflected by these objects: some of these echoes are received back by the LiDAR block 13 and captured by the sensors. Based upon the measures by the sensors of the received echoes, the LiDAR block 13 is adapted to measure the travel time of the laser echoes and to calculate the distance from the source to each reflecting object and also the direction of this object compared to a reference axis, for example the longitudinal axis of the train front. A 3D picture of the objects in the field of view of the LIDAR block 13 can thus be obtained at each time period T2 (for example T2 is in the range from 50ms to 3s).

[0037] The laser source is installed -and the emission of the LIDAR is tuned -in such a way that the LIDAR field of view centerline coincides the track centerline when the track ahead of the train is straight.

[0038] Scan speed of the LIDAR block 13 influences the number of points and echoes that are measured. The choice of optics and of scanner greatly influences the resolution and range of the LiDAR system. The range of the LIDAR is the length of the area in front of the train that can be monitored through the laser waves.

[0039] The database 11 stores data disclosing track network topology, enabling the knowledge of the 3D coordinates of each point of the rail network track or at least an accurate approximation.

[0040] For example, each track being represented by the median axis between the right and left rails of the track: -the median axis having been segmented into successive portions, each portion has been modeled by a segment; the succession of segments is stored in the database for example by chaining their identifier, the form and 3D geographical coordinates end points are known and stored in the database 11; and/or -the database 11 stores the 3D coordinates of a set of points along the median axis, the succession of this points being indicated in the database 11 for example by chaining their identifier and the distance between each point and its next point being equal or less than a predetermined distance d; for example d is in the range from 0.5m to 4m); d is for example chosen equal to 1 meter (m) ; for example, this embodiment is considered hereafter: figure 6 shows in a given plan P perpendicular to a track plane, the point 20 in the median axis between rail 21 and rail 22 of the track.

[0041] The database 11 stores also data defining a train clearance envelop all along each track. The determination of these data are now described referring to Figure 2, showing steps implemented in an embodiment of an obstacle detection method 100 according to the invention.

[0042] Considering a given plan P perpendicular to a track segment. The train clearance plate <in French "gabarit plan"> is the finite surface, referenced by S_clsd, made of -or containing -any point in said given plan P, that intersects the train while the train, from its beginning until its end, goes through this plane over the track.

[0043] In an embodiment, the considered surface is modeled by a simple geometrical form and/or is taken a little larger than the outline of the set of intersection points in said given plan. The margin between the considered surface and the intersection point outline is given by the infrastructure owner to guarantee the train clearance from construction phase already..

[0044] In the considered embodiment represented in figure 6, the train clearance outline 30 shown in the plan P, such modelized, is defined by the successive segments S30_1S30_2, S302S303, S303S304, S304S305, 53055306, 53065301 between the 6 vertices 53015 5302, 5303, 530_4, S30_5 et S30_6. The modeled surface S_clsd corresponds to the hatched area delimited by these segments.

[0045] The hereabove mentioned margin has to be defined in such a way for example that when train 1 enters a tunnel,the tunnel walls are outside the surface S_clsd or that known objects close to the train clearance envelope, e.g. platform, are not detected as obstacle in the steps 100_3 described later (in order to limit obstacle detection events that are not relevant).

[0046] Of course, the number of vertices defining the train clearance plate can be chosen different from 6.

[0047] In a preliminary step 100_1 as it appears in figure 2, the definition data, including its 3D geographical coordinates, of the global volume occupied by the train when it circulates along the whole track (from the departure point to the final destination point) is determined based upon the topography data of the track and upon the clearance plate 30 and the definition data of the determined volume, including its 3D geographical coordinates, are stored in the database 11.

[0048] In an embodiment, to that matter, the coordinates of the train clearance envelop 31 delimiting said volume are determined by deterministic algorithms and stored.

[0049] For example, as illustrated in figure 7, in order to limit the size of the stored data, a calculation is performed of the 3D coordinates of the vertices of the clearance plate 30 when positioned to each of the set of points 20, perpendicular the segment defined by said point and the next point of the set of point according to the direction of the train. This results in a sampling of the train clearance envelop. The envelop is obtained by linking each point of the clearance plate outline 30 to the corresponding point in the next clearance plate (for example, by linking the point S304 of a clearance plate to the point S304 of the next clearance plate).

[0050] In figure 7, the global volume, Vol, occupied by the train 1 is grey tint.

[0051] The absolute positions of the vertices of all these sets of vertices (1 set each d meters, each set comprising in the considered case 6 vertices) are stored in the database 11.

[0052] Preliminary step 100_1 is performed once regarding each train in regard to each track, before operational use of the obstacle detection solution defined by the steps 100_2, 100_3 and 100_4. Depending embodiments, step 100_1 is performed by the obstacle detection block 14 in the train 1 or by an electronic bloc of determination of the clearance envelops in a central system of the track network outside the train 1.

[0053] In an embodiment, in order to comply with for different train attitudes, e.g. on straight track or on curve with lateral profile, sensor (lidar) lever arms are taken into consideration when transforming the clearance outline 30 onto the LiDAR coordinates in order to determine if the reflected points are within the clearance (in step 100_3).

[0054] The obstacle detection block 14 is adapted to detect if obstacles are present on the trajectory of the train and to trigger actions in case of such a detection of present obstacles occurs, as detailed hereafter.

[0055] Referring again to figure 2 shows steps implemented by the processing module 10 in an embodiment of an obstacle detection method 100 according to the invention. The steps 100_2 to 100_3 here below are iterated every period T (for example T is in the range from 50ms to 3s, depending on the obstacle detection block 14 and necessity): [0056] In a step 100_2, when the train is moving along the track defined by the track rails 21, 22, at a given obstacle detection time t (period T), sub-steps 100_21 and 100_22 are performed.

[0057] In a sub-step 100_21, the obstacle detection block 14 gets from the LIDAR block 13 the more recent calculated 3D picture of the currently scanned area. The LiDAR point cloud is sliced into frames by the LIDAR block 13. The integration time depends on the required frame rate.

[0058] In a sub-step 100_22 parallel to sub-step 100_21, the obstacle detection block 14 gets from the location block 12 the last determined position of the train 1.

[0059] In a step 100_3 achieved by the obstacle detection block 14, the coordinates of the clearance envelop in front of the train are determined from the database 11 based upon the determined train position (and of the knowledge of the moving direction on the track) and transformed into LiDAR coordinates (in order to be in the same and one referential). The length of the considered clearance envelop (for example corresponding to a fixed number 31 of successive train clearance plates 30) depends on the detection range requirement (depending on needed braking distance and on LIDAR range). Then the obstacle detection block 14 determines if the LIDAR detected objects are inside or outside the volume delimited by the considered clearance envelop [0060] In case at least one object is detected inside, then in a step 100_4, an emergency action is triggered by the obstacle detection bloc 14: an alarm signal is generated, and/or an emergency/service braking is triggered (for example in case the obstacle is within the braking distance given the current speed of the train) ...To this end, the obstacle detection block 14 interfaces with signaling system, rolling stock, or provide alerts to the driver.

[0061] Applying the method according to the invention, and referring to figure 3, the detection of the object El, outside the volume described by the succession 31 of clearance plates 30 will not lead to an emergency action whereas the object E2, which is inside the volume, will cause an emergency action.

[0062] In an embodiment, considering the points of the generated LIDAR 3D picture, one or several of the followings are performed to determine if some of these points pertain to an object being really an obstacle needing the triggering of step 100_4: - a minimal size (s) of object (or of the part of the object inside the volume) is predefined: the block detection estimates a size of the detected object based upon the LIDAR picture and if the size is inferior to s, the detected object is not considered as a relevant obstacle and no emergency action is triggered; - the performance of the LiDAR: number of points n with respect to the LIDAR range (r) and minimal required obstacle size s: i.e. theoretically, at range r, for an object of size s, n laser rays should be received, and as a result reflected, as n reflected points (but in reality, this is not 100% guaranteed; if the LiDAR supplier is able to give a percentage (with sufficient proof), e.g. at least p of n can be reflected and shown on LiDAR image, we are able to know at range r, for a object of size s, we can have at least np points on the LiDAR image).

- the failure mode of the LiDAR: percentage (p) of the points (relative to a one detected object of size s) reflected and perceived with certain integrity level (i).

[0063] Assuming the target is to detect obstacle with more than a certain size of intrusion in the train clearance, for example, the following algorithm is used: - the object point of 3D LIDAR picture inside the volume that is nearest of the train (point 1 of the 3D LIDAR picture) within the considered clearance envelop is firstly selected; the number of the points (m) inside the envelope in the vicinity (i.e. within the distance of s) of the selected nearest point is determined; if m+l< np: point 1 is not regarded as pertaining to a relevant obstacle: point 1 is discarded, and the next nearest point is then considered by the algorithm. Otherwise, it is regarded as belonging to a relevant obstacle.

[0064] It must be noted that: -p depends on r and i and the specific LiDAR used; * n depends on r.

[0065] This arrangement enables to avoid "dust" or noise. To avoid machine learning, clustering of LiDAR point cloud is avoided. A threshold of the number of the reflected point is defined to identify if it is an obstacle or interference (e.g. dust) [0066] Handling of divergence switch ahead [0067] In an embodiment, in case there is a divergence switch ahead, both tracks will be scanned for detecting obstacles according to the invention. As shown in figure 5, the train 1 passes from a track (rails 21, 22) to another part of track which can possibly be a first track (21_1, 22_1) or a second one (21_2, 22_2), The two alternative portions are known in the database 11; thus the train clearance profile of both tracks after a switch is taken into account by the obstacle detection method: both tracks are detected for obstacles. Because if the train positions itself autonomously (not depending on wayside equipment), it is unknown which track the train will take after the switch. Also, when the train is on the switch, the position still can be ambiguous. After the train passes the switch, the train position is known, then only one track is further detected for obstacles. This makes the invention working with autonomous train positioning, without depending on the wayside information.

[0068] Train position determination by the location block 10 corresponds in an embodiment to a required safety level.

[0069] In an embodiment, information contained the data base 11 corresponds to a required accuracy and integrity level.

[0070] Safety demonstration can be performed through following options: if i is known, the detection integrity can be calculated.

-if i is unknown the LiDAR behavior can be monitored through the comparison of the LiDAR detection and known landmark information.

-redundancy of LIDAR sensors.

[0071] In an embodiment, the processing device 10 includes a microprocessor and a memory comprising instructions which, when executed by the microprocessor, implement one or several of the steps 100_2 to 100_4. Alternatively at least some of the steps can be implemented by dedicated hardware, typically a digital integrated circuit, either specific (ASIC) or based on programmable logic (e.g. FPGA/Field Programmable Gate Array).

[0072] The obstacle detected according to the invention as described hereabove can be train, human or any other objects.

[0073] Due to the constraints of LiDARs performance, the wider the FoV (field of view) is, the lower the point density is. Therefore, it may not be possible to find one single LiDAR to cover both the range and the width. Therefore, multiple LiDARs can be used. In an embodiment, two LiDAR are used or more, instead of only one, for long distance (Field of View FoV1) and short distance (Field of View FoV2) respectively to cover the whole range, as shown in figure 4. The benefits include to cover both width and range and that the FoV overlapping space can be detected with both/all the LiDARs to provide independent chains of detection to facilitate safety demonstration.

[0074] The invention provides a safety certifiable solution for railway obstacle detection, taking train speed, obstacle size, and LiDAR performance into consideration.

[0075] The solution is valuable for manual driving and autonomous driving mode. [0076] Lighting conditions does not affect the detection results.

[0077] The solution enables to detect object of certain size within defined range, depending on the performance of the LiDAR (range, point density, etc.). In an embodiment, the LiDAR range covers the braking distance for corresponding speed.

[0078] When a train is in manual driving mode, based on the speed restriction, considering the worst case, the emergency braking distance is defined as the range of obstacle detection and thus the range of the LIDAR block. This is to ensure the train is able to stop in front of the obstacle.

[0079] If the speed restriction is 5km/h, e.g. in the scenario of train coupling, the braking distance can be 10m.

[0080] If the speed restriction is 15km/h, e.g. Restricted Manual mode speed restriction imposed by some metro operators, the braking distance can be 40m.

[0081] If the speed restriction is 25km/h, e.g. Restricted Manual mode speed restriction imposed by some metro operators, the braking distance can be 80m.

[0082] If the speed restriction is 40km/h, e.g. Staff Responsible mode speed restriction imposed by ETCS or other mainline signal systems, the braking distance can be 170m.

[0083] Only deterministic algorithms are used according to embodiments of the invention (no need of machine learning or deep leaning) which makes safety demonstration possible here demonstration means a process of proving the safety level (hazard rate can be calculated)).

[0084] Failure modes of the LiDAR are defined for the specific LiDAR by the LiDAR block supplier. If not available, LiDAR monitoring functions can be implemented in the processing device 10.

[0085] In the particular embodiment described above with reference to Figure 6, the train clearance plate was estimated to be equal to a hexagon, with vertices S301, S302, S303, S304, S30_5 and S30_6. In one embodiment, the hexagon is defined by a set of rectangles. For example, with reference to Figure 6, the train clearance plate is estimated equal to that of two rectangles, one defined by the vertices 530_1, 530_2, 530_6 and 530_6and the other defined by the vertices 5303, 5304, 5'303 et 5'304, where 3'303 is the orthogonal projection of S303 on the segment S30_2S30_6 and S'30_4 is the orthogonal projection of S30_4 on this segment 53025306. The geographical coordinates of the clearance envelope are thus obtained as a function of this clearance plate and the trajectory defined by the railway track considered. The advantage conferred is to simplify the calculation load of the algorithmic treatments for calculating the envelope and for detecting obstacles according to the invention, which use the train clearance. If it is necessary to estimate the clearance more precisely, the clearance plate will be estimated by three rectangles or more rectangles.

[0086] Considering train attitude [0087] In one embodiment, the attitude of the train is also taken into account, as detailed hereabove.

[0088] As is known, the attitude information includes the pitch (slope or "pitch" in English) as illustrated schematically in Figure 8, the roll (lateral profile or "roll" in English) as illustrated schematically in Figure 9, and the yaw (cap, in English: "yaw" or "heading") as schematically illustrated in figure 10.

[0089] The train attitude depends on the track attitude.

[0090] In this embodiment, the database 11 stores at any point on the track whose 3D coordinates are stored (here the points 20), the attitude of the track at this point.

[0091] The attitude information is for example expressed, at each point considered on the track, relative to a North axis X, to an East axis Y and a vertical axis Z (i.e. a local tangent coordinate system of type NED).

[0092] Each train car is carried by two bogies. A bogie is a train car underneath structure to which axles (hence, wheels) are attached through bearings. Such a bogy is illustrated in Fig 12.

[0093] We consider the two bogies used to carry the front car of the train (i.e. the locomotive) : an example of a configuration of these two bogies are schematically represented in Fig 11 in top view, the front car body being represented by the grey rectangle (in other configurations, bogies are at the connection of two successive train cars). The sensor(s) (lidar) used for the invention method are installed on this front car (thus, in reality, each end of the car is equipped with the sensors and can be alternatively the front car, depending on the travel direction).

[0094] We consider the center point of each of these two bogies (i.e. the pivot of the bogy turning) carrying the front car: that is, with reference to Figure 11, the center points 81 and 82. We now consider the segment 83 of end points 81 and 82. The segment 83 remains rigid along with the front car body and has the same attitude as the front car body.

[0095] The dotted line 84 is the central line of the track on which the points 20 defined in the database 11 are located.

[0096] The relative position of the Lidar sensor fixed on the front train car with respect to the body of the front train car is known: with reference to Figure 12, considering as the origin 90 of a referring coordinates system, the point 81 or 82, the three-dimensional distance (considering each of the axes Xw, Yw, Zw) of the sensor with respect to the origin 90 is measured (in the form of-a-lever arms), in a preliminary step, for example during sensor installation.

[0097] In this embodiment, the obstacle detection block 10 determines the position of the segment 83 (i.e. of its ends 81, 82), then it calculates the attitude of the segment 83 as a function of the attitude of the track (stored in the database 11) in this position; then the obstacle detection block 10 calculates the position and the attitude of the sensor from the position and the attitude of the segment 83, and of the lever arms measured. [ [0098] Once the position and attitude of the sensor have been calculated: -the detection space (train clearance delimited by the envelope of the train) is converted into lidar coordinates (in step 100_3) as a function of the position and attitude of the sensor thus calculated.

[0099] A slight change in attitude of the sensor can have a significant impact on detection, in particular over long distances. With this method, detection accuracy is improved, especially for longer distances.

[0100] The invention has been disclosed hereabove using a LIDAR block. Other technology can be used instead of LIDAR, for example RADAR or SONAR technology, or any suitable technology using detection of echoes of waves generated aboard the train.

[0101] The invention has been disclosed hereabove regarding a train, but is more generally usable relative to any machine movable along any trajectory of a set of known trajectories, such machine being thus for example a metro, a tramway, a boat, a plane, a drone, with or without automatic and autonomous driving.

Claims (13)

1. CLAIMS1. Method for detecting obstacles (El, E2) implemented by an electronic obstacle detection device (10) onboard a machine (1) adapted to move on at least one predetermined path (21,22) ; said obstacle detection device (10) including: a database (11) storing definition data, including 3D geographical coordinates, of the global volume (31) occupied by the machine (1) when circulating along the predetermined path; a tele-detection block (13) adapted for emitting waves towards the path in front of the machine, for receiving echoes of emitted waves from at least one object and to calculate from said waves and echoes, direction and distance, relative to the machine (1), of said object; said method comprising the steps of: a/ determining the current 3D position of the machine (1) ; - b/ using the tele-detection block (13), determining direction and distance relative to the machine (1) of at least one currently detected object; - c/ as a function of the determined current 3D position, of the determined current direction and distance of the at least one detected object, determining, based upon said stored 3D geographical coordinates of the volume, if the at least one detected object is located inside said volume; d/ in case the at least one detected object is determined located inside said volume, triggering an emergency action.
2. 2. Method for detecting obstacles (El, E2) according to claim 1, wherein, in case at least one detected object is determined located inside said volume (31), an emergency action is triggered depending on the result of at least one additional step among: - estimating a size of the detected object based upon the received echoes, and verifying said estimated size is greater than a threshold size; - verifying the performance of the tele-detection block (13) based upon a number of points n with respect to the range r of the tele-detection and threshold size s of objects.
3. 3. Method for detecting obstacles according to claim 1 or 2, wherein, in case at least one detected object is determined located inside said volume (31), the following steps are achieved before triggering any emergency action: -i/ the object point inside the volume that is nearest of the machine (1) within the considered clearance envelop is firstly selected, and named point1; -ii/ the number m of the points inside the volume within a predefined distance s of the selected nearest point is determined; -iii/ if m+1< n.p: point1 is not regarded as pertaining to a relevant obstacle: point1 is discarded, no emergency action is triggered, and the next nearest point is then considered by the algorithm iterating from step i; n being the predefined number of theoretically reflected points and p being the percentage of the points reflected and perceived with given integrity level; otherwise if m+1a. n.p, point1 is regarded as belonging to a relevant obstacle and emergency action is triggered.
4. 4. Method for detecting obstacles according to any one of the preceding claims, wherein the global volume (31) occupied by the machine (1) is defined as a function of a set of rectangles defining a machine (1) clearance plate.
5. 5. Method for detecting obstacles according to any one of the preceding claims, wherein said database stores the path attitude along said predetermined path and the method comprises the following steps: -determining position and attitude of the tele-detection block (13) as a function of at least the current 3D position of the machine (1) and of the stored path attitude corresponding to the current 3D position of the machine; and said determination, in step c, of if the at least one detected object is located inside said volume is achieved furthermore as a function of said determined position and attitude of the tele-detection block (13).
6. 6. Method for detecting obstacles according to claim 5, wherein the machine (1) is a train having train cars and the method comprises the following steps: attitude and position of a segment between two bogies carrying the first train car are calculated; -tele-detection block position and attitude being determined as a function of said calculated segment attitude and position.
7. 7. Method for detecting obstacles according to any one of the preceding claims, wherein in case the path upon which the machine (1) moves is further going to separate into at least two alternative paths (21_1, 22_1, 21_2, 22_2), the database (11) storing definition data, including 3D geographical coordinates, of the global volume occupied by the machine (1) when circulating along each of the alternative path: -the situation of alternative paths is detected based upon of the determined current 3D position and the 3D geographical coordinates of the two global volume occupied by the machine when circulating along the at least two alternative paths; and -the steps c and d are then performed regarding the at least two alternatives paths.
8. 8. Method for detecting obstacles according to any one of the preceding claims, wherein the machine (1) is a train and the predetermined path (21, 22) is a track.
9. 9. Computer program, adapted to be stored in the memory of a obstacle detection device (10) further comprising a microcomputer, said computer program including instructions which, when executed on the microcomputer, implement the steps of a method according to any one of the preceding claims.
10. 10. Obstacle detection device (10) adapted to be onboard a machine (1) adapted to move on at least one predetermined path (21, 22) ; said obstacle detection device (10) including: a database (11) storing definition data, including 3D geographical coordinates, of the global volume (31)occupied by the machine when circulating along the predetermined path; a tele-detection block (13) adapted for emitting waves towards the path in front of the machine (1), for receiving echoes of emitted waves from at least one object and to calculate from said waves and echoes, direction and distance, relative to the machine, of said object; said method being adapted to implement the following operations of: a/ determining the current 3D position of the machine (1) ; b/ using the tele-detection block (13), determining direction and distance relative to the train of at least one currently detected object; c/ as a function of the determined current 3D position, of the determined current direction and distance of the at least one detected object, determining, based upon said stored 3D geographical coordinates of the volume (31), if the at least one detected object is located inside said volume; d/ in case the at least one detected object is determined located inside said volume, triggering an emergency action.
11. 11. Obstacle detection device (10) according to claim 10, adapted, in case at least one detected object is determined located inside said volume (31), to trigger an emergency action depending on the result of at least one additional operation implemented by the device (10) among: estimating a size of the detected object based upon the received echoes, and verifying said estimated size is greater than a threshold size; verifying the performance of the tele-detection block based upon a number of points n with respect to the range r of the tele-detection and threshold size s of objects.
12. 12. Obstacle detection device (10) according to claim 10 or 11, adapted, in case at least one detected object is determined located inside said volume (31), to achieve the following operations before triggering any emergency action: -i/ the object point inside the volume that is nearest of the machine (1) within the considered clearance envelop is firstly selected, and named point1; -ii/ the number m of the points inside the volume within a predefined distance s of the selected nearest point is determined; -iii/ if m+1< n.p: point1 is not regarded as pertaining to a relevant obstacle: point1 is discarded, no emergency action is triggered, and the next nearest point is then considered by the algorithm iterating from operation i; n being the predefined number of theoretically reflected points and p being the percentage of the points reflected and perceived with given integrity level; otherwise if m+Th n.p, point1 is regarded as belonging to a relevant obstacle and emergency action is triggered.
13. 13. Obstacle detection device (10) according to any one of the preceding claims 10 to 12, wherein the database (11) storing definition data, including 3D geographical coordinates, of the global volume occupied by the machine when circulating along each of the alternative path, the obstacle detection device (10) being adapted, in case the path (21, 22) upon which the machine moves is further going to separate into at least two alternative paths (21_1, 22_1, 21_2, 22_2), to detect the situation of alternative paths based upon of the determined current 3D position and the 3D geographical coordinates of the two global volume (31) occupied by the machine when circulating along the at least two alternative paths; and to perform then the operations c and d regarding the at least two alternatives paths.