Path determination device, path determination method, and non-transitory computer-readable medium

Abstract

A GP acquisition unit (1) acquires a global path (GP) in which a plurality of relay points are provided. A relay point selection unit (2) selects a relay point (WPj) such that there is no duplication. A path creation unit (3) creates a plurality of paths (PA1 to PAN) from the current position of a vehicle to the relay point (WPj). A path selection unit (4) selects one path (PAk) such that there is no duplication. A score calculation unit (5) calculates a score (SAk) of the path (PAk). A similarity determination unit (6) determines the similarity between the plurality of paths (PA1 to PAN) and the previous path. A score adjustment unit (7) adjusts the score (SAk) of the path (PAk) when there is similarity. An optimal path selection unit (8) selects an optimal path (Pj) to the relay point (WPj) on the basis of the score (SAk). A local path determination unit (9) determines the local path (LP) on the basis of the respective scores of the plurality of optimal paths determined for the plurality of relay points.

Description

Route determination device, route determination method, and non-transitory computer-readable medium

[0001] The present disclosure relates to a route determination device, a route determination method, and a non-transitory computer-readable medium.

[0002] In recent years, there has been progress in the development of automated vehicle operation technologies. In particular, various technologies for automatically creating routes for vehicles have been proposed (Patent Documents 1 to 5).

[0003] For example, Patent Document 1 proposes a method for creating the shortest possible route to a relay point set on a global route, taking into consideration factors such as the presence of obstacles.

[0004] US Patent Application Publication No. 2019 / 098082 JP 2022-157259 A International Publication No. 2021 / 181652 JP 2020-086588 A International Publication No. 2019 / 098082

[0005] Generally, when considering efficient vehicle operation, it is desirable to avoid sudden changes in speed or direction of travel. However, if a route with a shorter distance is selected preferentially, it is expected that sudden changes in speed or direction of travel will occur. Furthermore, if too sudden a change in speed or direction of travel is required, it may cause disruption to the operation of the vehicle, and the vehicle may even be forced to stop.

[0006] The present disclosure has been made in consideration of the above circumstances, and aims to automatically determine a route that allows a vehicle to travel smoothly.

[0007] A route determination device according to one aspect of the present disclosure includes a global route acquisition unit that acquires a global route having multiple relay points, a relay point selection unit that selects one relay point from the multiple relay points so that the relay points do not overlap, a route creation unit that creates multiple routes from the current position of a vehicle to the selected relay point, a route selection unit that selects one route from the multiple created routes so that the routes do not overlap, a score calculation unit that calculates a score for evaluating the selected route based at least on its length, a similarity determination unit that determines the similarity between the multiple routes and a previous route that is a route to the selected relay point that was previously determined, a score adjustment unit that adjusts the calculated score if each of the multiple routes is similar to the previous route so that the evaluation of each of the multiple routes is higher, an optimal route selection unit that selects the route with the highest evaluation as the optimal route to the selected relay point based on the scores of each of the multiple optimal routes determined for the multiple relay points, and determines the route with the highest evaluation as the local route based on the scores of each of the multiple optimal routes determined for the multiple relay points.

[0008] A route determination method according to one aspect of the present disclosure includes obtaining a global route having multiple relay points, selecting one relay point from the multiple relay points without overlapping each other, creating multiple routes from the current position of a vehicle to the selected relay point, selecting one route from the multiple routes created without overlapping each other, calculating a score for evaluating the selected route based at least on its length, determining the similarity between the multiple routes and a previous route that is a route to the selected relay point that was previously determined, and if each of the multiple routes is similar to the previous route, adjusting the calculated score so that each of the multiple routes is evaluated higher, selecting the route with the highest evaluation based on the score for each of the multiple routes as an optimal route to the selected relay point, and determining the route with the highest evaluation as a local route based on the score for each of the multiple optimal routes determined for the multiple relay points.

[0009] A non-transitory computer-readable medium storing a program that is one aspect of the present disclosure causes a computer to perform the following operations: acquiring a global route having multiple relay points; selecting one relay point from the multiple relay points so that the relay points do not overlap; creating multiple routes from the current position of a vehicle to the selected relay point; selecting one route from the multiple created routes so that the routes do not overlap; calculating a score for evaluating the selected route based at least on its length; determining the similarity between the multiple routes and a previous route that is a route to the selected relay point that was previously determined; if each of the multiple routes is similar to the previous route, adjusting the calculated score so that each of the multiple routes is evaluated higher; selecting the route with the highest evaluation as an optimal route to the selected relay point based on the scores of each of the multiple routes; and determining the route with the highest evaluation as a local route based on the scores of each of the multiple optimal routes determined for the multiple relay points.

[0010] According to the present disclosure, a route that allows a vehicle to travel smoothly can be automatically determined.

[0011] FIG. 1 is a diagram schematically illustrating a configuration of a route determination system according to a first embodiment. FIG. 2 is a diagram illustrating an example of a route. FIG. 3 is a diagram schematically illustrating a configuration of a local planning unit according to a first embodiment. FIG. 4 is a flowchart of a local route determination process in the local planning unit according to the first embodiment. FIG. 5 is a diagram illustrating an overview of similarity determination. FIG. 6 is a diagram schematically illustrating a configuration of a local planning unit according to a second embodiment. FIG. 7 is a flowchart of a local route determination process in the local planning unit according to the second embodiment. FIG. 8 is a diagram illustrating an overview of route selection based on scores. FIG. 9 is a diagram illustrating an overview of score updating based on costs. FIG. 10 is a diagram schematically illustrating a configuration of a computer as an example of a hardware configuration.

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same elements are designated by the same reference numerals, and redundant explanations will be omitted as necessary.

[0013] Hereinafter, when we refer to one embodiment, it means that it can be applied to any one of the embodiments described below or a combination of two or more embodiments, and that the application is not limited to a specific embodiment.

[0014] First Embodiment A route determination system according to a first embodiment will be described. FIG. 1 schematically shows the configuration of a route determination system 100 according to the first embodiment. The route determination system 100 is used to determine a route for a vehicle to travel from a departure point to a destination. The route determination system 100 also plans vehicle control so that the vehicle faces a desired direction when it arrives at the destination, in addition to traveling from the departure point to the destination. For example, it is desirable for a forklift in a warehouse not only to reach the shelf where the product to be removed is stored, but also to face the shelf to increase the efficiency of movement. In this way, when controlling the operation of a vehicle, the travel route and the vehicle's orientation are controlled together.

[0015] The configuration of the route determination system 100 will be described below. Fig. 1 shows a block diagram of the route determination system 100 according to one embodiment. The route determination system 100 includes a position and orientation acquisition unit 110, a control unit 130, a sensor 150, and a plan creation unit 160. The route determination system 100 is also provided with destination pose information 120 and map information 140 as appropriate.

[0016] The position and attitude acquisition unit 110 acquires latest pose information CP, which is information indicating the position and attitude of the vehicle at its current location. Hereinafter, the pose information is given as a combination of the vehicle's position and the vehicle's attitude at that position, specifically, the vehicle's orientation. In one embodiment, the pose information may further include the vehicle's speed at the time the pose information was acquired. In one embodiment, the position and attitude acquisition unit 110 may include at least one of various sensors attached to the vehicle, such as an accelerometer or a gyroscope. In one embodiment, the position and attitude acquisition unit 110 may be implemented by a device external to the vehicle that can detect the vehicle's position and attitude. For example, in one embodiment, the vehicle may be monitored by a remote sensor such as a camera or LiDAR (Light Detection and Ranging), and the vehicle's pose information may be provided by a processor connected to the vehicle. The position and attitude acquisition unit 110 independently outputs the latest pose information CP to the control unit 130 and the plan creation unit 160 via wireless communication, wired communication, or a combination thereof.

[0017] The destination pose information DP holds the destination position and attitude of the vehicle, i.e., the desired orientation of the vehicle when it reaches the destination. In one embodiment, the destination pose information DP may be provided by a user via various input / output means, such as a graphical user interface (GUI). In one embodiment, the destination pose information DP is provided by an external device, such as a processor, configured to determine the vehicle's operation. The destination pose information DP is input to the plan creation unit 160 via wireless communication means, wired communication means, or a combination thereof.

[0018] The control unit 130 outputs commands COM to the vehicle to command both the speed and direction of travel of the vehicle to control the vehicle's travel. In one embodiment, the control unit 130 may be configured as a part of the vehicle. In one embodiment, the control unit 130 may be provided external to the vehicle. In one embodiment, the control unit 130 may be a cloud-based control unit implemented using one or more servers. The control unit 130 receives the local path LP and waypoints from the plan creation unit 160 and the latest pose information CP from the position and attitude acquisition unit 110 via wireless communication means, wired communication means, or a combination thereof. The control unit 130 determines what operation the vehicle should perform based on the received local path LP and waypoints. In one embodiment, the control unit 130 may output the commands COM to the vehicle via wireless communication means, wired communication means, or a combination thereof.

[0019] The map information 140 includes information indicating obstacles in an area where the vehicle is scheduled to travel. In one embodiment, the map information 140 may be stored in a storage device installed in the vehicle. In one embodiment, the map information 140 may be stored in a storage device external to the vehicle, such as a cloud-based storage device. The map information 140 includes information regarding the locations of known obstacles in an area where the vehicle is scheduled to travel. In one embodiment, the map information 140 may be updated as needed when a new obstacle is detected or when the location of a known obstacle changes. In one embodiment, the detection of a new obstacle or a change in the location of a known obstacle may be provided by various devices that acquire information about the controlled vehicle, other vehicles, and obstacles. In one embodiment, the map information 140 may include information regarding routes on which the vehicle is navigable, such as roads, aisles in a warehouse, or corridors in a building. In one embodiment, the map information 140 may further include information related to the locations of vehicles other than the controlled vehicle. The obstacle information and route information included in the map information 140 are transmitted to the plan creation unit 160 by wireless communication means, wired communication means, or a combination thereof. In one embodiment, the map information 140 may be provided to the control unit 130 by wireless communication means, wired communication means, or a combination thereof.

[0020] The sensor 150 detects obstacles near the vehicle. In one embodiment, the sensor 150 may be configured as a single sensor. In one embodiment, the sensor 150 may be configured by multiple sensors. In one embodiment, the sensor 150 may be configured as a combination of different types of sensors. In one embodiment, the sensor 150 may be provided inside the vehicle. In one embodiment, the sensor 150 may be provided outside the vehicle. In one embodiment, the sensor 150 may be configured as an integrated part of the position and attitude acquisition unit 110. In one embodiment, the sensor 150 may be configured by various sensors such as a camera or LiDAR. The sensor 150 provides information about the surroundings of the vehicle, such as various data such as point cloud data or laser scan data, to the plan creation unit 160 via wireless communication means, wired communication means, or a combination thereof.

[0021] The plan creation unit 160 receives the latest pose information CP from the position and attitude acquisition unit 110. The plan creation unit 160 also receives destination pose information 120 and obstacle information and route positions included in the map information 140. Hereinafter, the received destination pose information 120 will be referred to as destination pose information DP. Furthermore, the plan creation unit 160 receives information about the surroundings of the vehicle from the sensor 150. Using the received information, the plan creation unit 160 determines a local route, which is a route for the vehicle to travel between the current location and the destination, and intermediate points. The plan creation unit 160 outputs the determined local route and intermediate points to the control unit 130 via wireless communication, wired communication, or a combination of wireless and wired communication. The plan creation unit 160 is implemented using one or more processors. In one embodiment, the plan creation unit 160 may be configured as a part of the vehicle. In one embodiment, the plan creation unit 160 may be provided external to the vehicle.

[0022] The plan creator 160 includes a global plan module 162 and a local plan module 165 .

[0023] The global planning module 162 determines global relay points between the current position and the target position at predetermined intervals. This interval may be, for example, every 100 milliseconds, or may be determined when the current position has moved 1 meter from the current position at the time the previous global relay point was determined. Hereinafter, the current position of the vehicle will be simply referred to as the current position, and the destination position will be referred to as the target position. The global planning module 162 includes a global cost map 163 and a global planning unit 164. The global cost map 163 determines which areas within the map the vehicle can travel between the current position and the target position. The global planning unit 164 determines relay points between the current position and the target position based on information received by the planning unit 160 and information received from the global cost map 163. In one embodiment, the global planning unit 164 may set relay points between the current position and the target position at regular intervals. In one embodiment, global planner 164 may set relay points at positions that are a predetermined distance from the positions of known obstacles between the current location and the target location. Global cost map 163 and global planner 164 are implemented by one or more processors. In one embodiment, global cost map 163 and global planner 164 may be implemented by the same processor. In one embodiment, global cost map 163 and global planner 164 may be implemented by separate processors.

[0024] The local planning module 165 determines a local route and intermediate points from the vehicle's current location to the intermediate points determined by the global planning module 162. The local planning module 165 includes a local cost map 166 and a local planner 167. The local cost map 166 determines which areas of the map are passable by the vehicle between the current location and the selected intermediate points. The local planner 167 determines a route from the current location to the selected intermediate points based on information received from the local cost map 166 and information received from the global planning module 162. The local planner 167 also sets intermediate points along the determined route. The local cost map 166 and the local planner 167 are implemented by one or more processors. In one embodiment, the local cost map 166 and the local planner 167 may be implemented by the same processor. In one embodiment, the local cost map 166 and the local planner 167 may be implemented by separate processors. In one embodiment, at least one of local cost map 166 and local planner 167 may be implemented by the same processor as at least one of global cost map 163 and global planner 164. In one embodiment, local cost map 166, local planner 167, global cost map 163, and global planner 164 may each be implemented by a separate processor.

[0025] 2 shows an example of a route. In one embodiment, route 200 is a global route provided by global planning module 162 to local planning module 165. Route 200 includes current pose information CP and destination pose information DP. Current pose information CP includes a current vehicle position 210 and a vehicle orientation 215 at current position 210. Destination pose information DP includes a target vehicle position 220 and a desired vehicle orientation 225 at target position 220. There are multiple intermediate points 205 between current position 210 and target position 220. Intermediate point 230 is a intermediate point selected from the multiple intermediate points 205 by local planning unit 167 to determine a local route starting from current position 210.

[0026] Based on the route 200, the local planner 167 determines a local route from the current position 210 to the relay point 230, in which a desired orientation 235 of the vehicle at the relay point 230 is set. The control unit 130 receives the local route and the relay points on the local route, and in response thereto, the control unit 130 creates a command COM for moving the vehicle from the current position 210 to the relay point 230, and outputs the created command COM to the vehicle.

[0027] As a method for creating a global route, for example, the method disclosed in Patent Document 1 or other known methods can be used as appropriate.

[0028] In this embodiment, when determining a local route, the previous processing history is referenced to create a local route that can ensure smooth vehicle operation. The following describes a route determination device 1000 that determines the local route. The route determination device 1000 corresponds to the local planner 167 described above.

[0029] 3 schematically shows the configuration of the route determination device 1000. The route determination device 1000 includes a global route acquisition unit 1, a relay point selection unit 2, a route creation unit 3, a route selection unit 4, a score calculation unit 5, a similarity determination unit 6, a score adjustment unit 7, an optimal route selection unit 8, and a local route determination unit 9. Note that, hereinafter, the global route acquisition unit will also be referred to as a GP acquisition unit.

[0030] The operation of each unit of the route determination device 1000 will be described with reference to a flowchart. Fig. 4 shows a flowchart of the process of determining a local route in the route determination device 1000.

[0031] Step ST1: The GP acquisition unit 1 reads a global route and a plurality of relay points set on the global route, for example, from the global plan module 162. In the following description, M is an integer equal to or greater than 2, and the number of relay points is M.

[0032] Step ST2: The relay point selection unit 2 selects M relay points WP １ ~WP Ｍ From one relay point WP ｊ In the following, the process for one selected relay point is repeatedly executed for each of the M relay points, and the j-th selected relay point is referred to as relay point WP ｊ Here, j is an integer between 1 and M. Here, M relay points WP １ ~WP Ｍ The relay points that have not yet been selected are selected as one relay point WP ｊ Select as.

[0033] Step ST3: The route generation unit 3 determines the relay point WP selected from the current position of the vehicle indicated by the latest pose information CP. ｊ Here, there are N routes PA １ ~PA Ｎ Here, N is an integer equal to or greater than 2. １ ~PA Ｎ is created as a Reeds-Shepp curve, for example.

[0034] Step ST4: The route selection unit 4 selects the route PA １ ~PA Ｎ In the following, the process for the selected route is performed on the N routes PA １ ~PA Ｎ The k-th selected path is designated as the selected path PA ｋ Here, k is an integer between 1 and N. Here, N routes PA １ ~PA ＮA route that has not yet been selected is referred to as a selected route PA ｋ Select as.

[0035] Step ST5: The score calculation unit 5 calculates the selected path PA ｋ Score SA for evaluating ｋ Here, the score SA ｋ As shown in the following equation, the selected path PA ｋ The length d ｋ Let's say.

[0036] Score S ｋ The calculation method of is not limited to Equation [1], and other calculation methods may be used. An example of calculation of the score according to one embodiment will be described. For example, it is assumed that the selected route is not suitable for the vehicle to travel due to the presence of an obstacle on the map on which the vehicle travels. Furthermore, if the number of times the vehicle changes direction on the selected route is large, the route length will be long, and the vehicle operation efficiency will be reduced.

[0037] Therefore, in order to select a route suitable for vehicle travel, a score may be calculated based on the following formula, as disclosed in Patent Document 1, for example. N ＲＳ is the number of vehicle turns in the route, and N ＲＳ The lower the score, the higher the SA. ｋ becomes smaller. ＲＳ is the length of the Reeds-Shepp curve. Ｇ is the length of the global path from the vehicle's current position to the intermediate point. ＲＳ / I Ｇ The smaller the score, SA ｋ becomes smaller. θ ｄｉｆｆ is the angular difference between the direction of the vehicle at the current position and the direction of the vehicle at the relay point, and θ ｄｉｆｆ The smaller the score SA ｋ becomes smaller. θ Ｈ is the angle between the direction of the vehicle at the current position and the line connecting the current position and the destination position. d is the straight-line distance between the current position of the vehicle and the relay point. The longer d is, the higher the score SA ｋ becomes smaller. ｏｐｔis the optimal Euclidean distance between the current position of the vehicle and the intermediate point. For example, d ｏｐｔ is determined appropriately depending on various parameters such as the density of obstacles in the map and the dimensions of the vehicle. ｄ is the distance to an obstacle near the relay point, and R ｄ The longer the score, the higher the SA ｋ In one embodiment, R ｄ may be the shortest distance between the created path and the obstacle. １ ~a ６ is an arbitrary coefficient.

[0038] As described above, the score used here is calculated so that the smaller the value, the higher the evaluation of the corresponding route. However, this is merely an example, and the score may be calculated so that the larger the value, the higher the evaluation of the corresponding route, if necessary.

[0039] Step ST6: The similarity determination unit 6 determines whether the selected path PA ｋ is the previous local route LP determined as the local route in the previous local route determination process. ＰＲＥ In this example, the similarity determination unit 6 compares the first three sequences among the multiple sequences that make up the so-called Reeds-Shepp curve, and determines that the first three sequences are similar if they are the same, and that the first three sequences are dissimilar if they are different.

[0040] An outline of the similarity determination is shown in FIG. 5. When showing routes in FIG. 5, R means a right turn, L means a left turn, S means a straight line, + means forward, and − means backward. In this example, the previous local route LP ＰＲＥis composed of four sequences: "L-R+S+R+". In this case, the first three sequences are "L-R+S+". In contrast, in the current process, three candidate routes are created. The first route is L-R+S+R+, the second route is L-R+S+L+, and the third route is L-R+L+R+. Of these three routes, the first and second routes have the same first three sequences as the route created previously. In this case, in the current process, the third route is not selected, and either the first or second route is selected.

[0041] Selected Route PA ｋ is the previous local route LP ＰＲＥ If the selected path PA is not similar to the selected path PA, the similarity determination unit 6 ｋ The route selector 4 then transmits a notification N1 indicating that the process for the route has been completed.

[0042] Selected Route PA ｋ is the previous local route LP ＰＲＥ If the result is similar to the above, the similarity determination unit 6 outputs the determination result D ｋ is transmitted to the score adjustment unit 7.

[0043] Step ST7: Selected path PA ｋ is the previous local route LP ＰＲＥ If the result is similar to the score, the score adjuster 7 determines the result D ｋ Depending on the score, SA ｋ For example, the score S ｋ is calculated by the following formula: In equation [3], μ is the selected path PA ｋ is the previous local route LP ＰＲＥ This is the discount rate for the score when the score is similar to the score of the original score, and can be any value greater than 0 and less than 1. Here, μ=0.1 as an example.

[0044] Note that the formula [3] is merely an example, and the selected path PA ｋ is the previous local route LP ＰＲＥ Score S when similar to ｋ The initial score SA ｋAny calculation may be performed to make the score SA smaller than ｋ Subtract a certain value from the score S ｋ Alternatively, for example, the score SA ｋ is multiplied by a coefficient greater than 0 and less than 1 to obtain the score S ｋ may be calculated.

[0045] When the score adjustment is completed, the score adjustment unit 7 selects the path PA ｋ The route selector 4 then transmits a notification N1 indicating that the process for the route has been completed.

[0046] Step ST8 Route PA ｋ When the process for the route PA is completed, the route selection unit 4 selects the route PA created in step ST3 in response to the reception of the notification N1. １ ~PA Ｎ Among them, the selected route PA ｋ It is determined whether there is a route that has not yet been selected. １ ~PA Ｎ Among them, the selected route PA ｋ If there is a route that has not yet been selected as the route, the process returns to step ST4.

[0047] Step ST9 Route PA １ ~PA Ｎ Among them, the selected route PA ｋ If there is no route that has not yet been selected as １ ~PA Ｎ For each of the scores SA １ ~SA Ｎ In response to this, the optimum route selection unit 8 calculates the score SA １ ~SA Ｎ Refer to Route PA １ ~PA Ｎ The route with the smallest score among the routes is called the intermediate point WP ｊ Optimal path P for ｊ Then, the optimum route selection unit 8 determines the relay point WP ｊ The relay point selector 2 then transmits a notification N2 indicating that the process for the item (1) has been completed.

[0048] Step ST10 Relay point WPｊ When the process for the relay points WP is completed, the relay point selection unit 2 selects the relay points WP in response to the reception of the notification N2. １ ~WP Ｍ It is determined whether there are any relay points that have not yet been selected among the M relay points WP １ ~WP Ｍ If there is any relay point that has not yet been selected among the M relay points WP, the process returns to step ST2. １ ~WP Ｍ If there is no relay point that has not been selected among the relay points WP １ ~WP Ｍ The local route determination unit 9 then transmits a notification N3 to the local route determination unit 9 indicating that the process for each of the above has been completed.

[0049] Step ST11 Relay point WP １ ~WP Ｍ When the process for each of the above is completed, the local route determination unit 9 determines the optimum route P １ ~P Ｍ The local route determination unit 9 determines the route having the smallest score as the local route LP for control. The local route determination unit 9 outputs the determined local route LP to the sensor 150. The determined local route LP is used as the previous local route LP in the next route determination process. ＰＲＥ It is referred to as.

[0050] By adjusting the calculated score for a route based on its similarity to the previous route, the route that is similar to the previous route is more likely to be selected as the optimal route. This effectively prevents sudden acceleration / deceleration and changes in direction of the vehicle, enabling smooth operation.

[0051] In the first embodiment, a configuration was described in which a route that allows a vehicle to travel smoothly for one relay point was selected based on the similarity with the previous local route. However, if there is an obstacle that hinders the vehicle's travel, it is necessary to set a route that avoids the obstacle. Therefore, the following describes route setting that takes into consideration the avoidance of the obstacle.

[0052] 6 is a schematic diagram showing a configuration of a route determination device 2000 according to the second embodiment. The route determination device 2000 has a configuration in which a score update unit 10 is added to the route determination device 1000.

[0053] The operation of each unit of the route determination device 2000 will be described. Fig. 7 shows a flowchart of the local route determination process in the route determination device 2000. In Fig. 7, step ST21 is added before step ST11, as compared to Fig. 4.

[0054] In this embodiment, in step ST5, for each of the multiple routes for each relay point, the score SA is calculated by taking into account the cost of the cost map through which the route passes, for example, taking into account the influence of an obstacle. ｋ The path determination device 2000 may obtain a cost map, for example, from the local cost map 166 as needed.

[0055] Specifically, in step ST5, first, the score calculation unit 5 calculates the route PA ｋ The maximum cost maxCA on the score map through which ｋ and average value aveCA ｋ Next, the score calculation unit 5 calculates the score SA by adding a term related to cost to the right side of the formula [1] as follows: ｋ Calculate. ρ is a coefficient indicating the contribution of cost to the score.

[0056] According to equation [4], the larger the cost, the smaller the score SA ｋ As a result, the score of the route close to the obstacle becomes large, and the intermediate point WP ｊ It can be seen that the probability of being determined as the optimal route to is low.

[0057] However, the score calculation here is merely an example, and the score may be calculated in the same manner as in the first embodiment.

[0058] In FIG. 8, the score SA ｋ8 shows an overview of route selection based on the above. On the cost map, a cost distribution is set within a predetermined distance from an obstacle OBS for the vehicle V, so that the value decreases as the distance from the obstacle OBS increases. In FIG. 8, contour lines indicating equal costs are displayed around the obstacle OBS, with the inner contour lines indicating higher costs.

[0059] In FIG. 8, the relay point WP ｊ Route PA towards １ ~PA ４ Assume that route PA is created. １ is the relay point WP ｊ It is the shortest route to the obstacle OBS, but it is the route closest to the obstacle OBS. ２ is the second shortest path to the relay point WP, and passes near the outer edge of the area around the obstacle OBS where the cost is set. ３ and P.A. ４ Although the route PA does not pass through the area around the obstacle OBS where the cost is set, １ and P.A. ２ This is a significantly longer route than

[0060] In this case, by using the score calculated by the formula [4], a route with the shortest possible distance and the lowest possible cost is selected. ２ is selected as the optimal route.

[0061] Next, step ST21 will be described. In step ST21, the score of the optimum route determined for each relay point may be adjusted taking into account the influence of obstacles. For example, on the cost map, the optimum route P ｊ The cost of the cost map through which the vector passes may be calculated, and the score may be adjusted according to the calculation result.

[0062] Specifically, the score update unit 10 ｊ The optimal path P determined for ｊ The maximum cost on the score map through which each of ｊ and average value aveC ｊNext, the score update unit 10 calculates the optimal route P １ ~P Ｍ Score S １ ~S Ｍ The calculated cost C １ ~C Ｍ The score is updated using the following equation [5]. ω is an arbitrary positive coefficient.

[0063] An overview of score updating based on cost is shown in Fig. 9. As in Fig. 8, a cost distribution is set on the cost map within a predetermined distance from the obstacle OBS so that the value decreases with increasing distance from the obstacle OBS.

[0064] In FIG. 9, two optimal paths P １ and P ２ The optimal route P ２ is the optimal path P １ In comparison with the optimal path P １ is a route that avoids the obstacle OBS. In this case, the optimal route P １ is the optimal path P ２ However, considering the smoothness of operation, the vehicle will take the optimal route P １ It is desirable to drive

[0065] In this case, by using the score calculated by the formula [5], the pre-adjustment score S ｊ In FIG. 9, by appropriately setting the coefficient ω, for example, the optimal route P １ is selected as the local path LP. This makes it possible to prevent an optimal path that is excessively close to the obstacle OBS from being selected as the local path.

[0066] As described above, according to this configuration, it is possible to realize smooth operation by suitably preventing sudden acceleration / deceleration and direction changes of the vehicle, as in embodiment 1. Furthermore, according to this configuration, it is also possible to select a route that avoids excessive proximity to obstacles, and it is possible to realize route determination that is more in line with the actual operation of the vehicle.

[0067] Other Embodiments The present invention is not limited to the above-described embodiment, and modifications can be made as appropriate without departing from the spirit of the present invention. For example, in the above-described embodiment, the first three sequences of the Reeds-Shepp curve are used to determine the similarity between two paths in step ST6. However, this is merely an example. For example, depending on the application, the similarity between two paths may be determined using any number of sequences, two or more, of the first two sequences of the Reeds-Shepp curve.

[0068] In the second embodiment, the maximum and average costs on the cost map through which the route passes are calculated, but the cost evaluation method is not limited to this. For example, an area occupied by the vehicle body may be set, and the maximum and average costs in the area swept by the vehicle body as the vehicle passes along the route may be calculated. In this case, the influence of the vehicle body can be taken into account, thereby improving the possibility of selecting a route that can more effectively avoid contact between the vehicle and obstacles.

[0069] In FIG. 5, the three routes are the previous local route LP ＰＲＥ Although the example of determining similarity by comparing with the local path LP is described above, this is merely an example. ＰＲＥ may be compared to determine the similarity.

[0070] 8, an example in which four routes are created has been described, but the number of routes to be created is not limited to four. In other words, any number of routes may be created.

[0071] 9, an example in which a local route is determined from two optimum routes is described, but this is merely an example, and a local route may be determined from any number of optimum routes.

[0072] The route determination system and route determination device can be realized using hardware resources such as a computer. FIG. 10 schematically shows the configuration of a computer 3000, which is an example of a hardware configuration. The computer 3000 can be configured as various types of computers, such as a dedicated computer or a personal computer (PC). However, the computer does not need to be a single physical computer; multiple computers may be used when performing distributed processing. As shown in FIG. 1, the computer 3000 has a CPU (Central Processing Unit) 1001, a ROM (Read Only Memory) 1002, and a RAM (Random Access Memory) 1003, which are interconnected via a bus 1004. Note that although an explanation of the OS software and other software required to operate the computer is omitted, it is assumed that the computer also has these software.

[0073] An input / output interface 1005 is connected to the bus 1004. To the input / output interface 1005, an input unit 1006, an output unit 1007, a communication unit 1008, and a storage unit 1009 are connected.

[0074] The input unit 1006 is composed of, for example, a keyboard, a mouse, a sensor, etc. The output unit 1007 is composed of, for example, a display device such as an LCD, and an audio output device such as headphones and speakers, etc. The communication unit 1008 is composed of, for example, a router, a terminal adapter, etc. The storage unit 1009 is composed of a storage device such as a hard disk, a flash memory, etc.

[0075] The CPU 1001 can perform various processes in accordance with various programs stored in the ROM 1002 or various programs loaded from the storage unit 1009 to the RAM 1003. In this embodiment, the CPU 1001 executes processes performed by the terminal devices T1 to T4, for example. A GPU (Graphics Processing Unit) may be provided separately from the CPU 1001, and, like the CPU 1001, executes various processes in accordance with various programs stored in the ROM 1002 or various programs loaded from the storage unit 1009 to the RAM 1003, for example, processes performed by the terminal devices T1 to T4 in this embodiment. Note that the GPU is suitable for performing routine processes in parallel, and by applying it to neural network processing, which will be described later, it is possible to improve the processing speed compared to the CPU 1001. The RAM 1003 also stores data necessary for the CPU 1001 and the GPU to execute various processes, as appropriate.

[0076] The communication unit 1008 is capable of two-way communication with a server via a network. The communication unit 1008 can transmit data provided by the CPU 1001 to the server, and output data received from the server to the CPU 1001, RAM 1003, storage unit 1009, etc. The communication unit 1008 may communicate with other devices using analog or digital signals. The storage unit 1009 is capable of exchanging data with the CPU 1001, and stores and erases information.

[0077] A drive 1010 may be connected to the input / output interface 1005 as needed. Storage media such as a magnetic disk 1011, an optical disk 1012, a flexible disk 1013, or a semiconductor memory 1014 may be appropriately mounted in the drive 1010. Computer programs read from each storage medium may be installed in the storage unit 1009 as needed. Furthermore, data required for the CPU 1001 to execute various processes, data obtained as a result of the processes of the CPU 1001, and the like may be stored in each storage medium as needed.

[0078] For example, the processes performed by the route determination system and route determination device according to the above-described embodiments may be realized by causing a computer to execute a program. Specifically, one or more programs containing instructions for causing a computer system to execute an algorithm related to the transmission signal processing or the reception signal processing may be created, and the programs may be supplied to a computer that realizes the route determination system and the route determination device.

[0079] These programs can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-RWs, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The programs may also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can supply the programs to a computer via wired communication paths such as electric wires and optical fibers, or via wireless communication paths.

[0080] Note that information such as the destination pose information 120 and the map information 140 in the above-described embodiment may be stored in a storage means such as the storage unit 1009 of the computer 3000 shown in Fig. 10 and read out as appropriate by the CPU 1001. Furthermore, information generated as a result of processing in the route determination system and route determination device, or information generated during the course of processing, may be stored in a storage means such as the storage unit 1009 as necessary and read out as appropriate by the CPU 1001.

[0081] REFERENCE SIGNS LIST 1 GP acquisition unit 2 Intermediate point selection unit 3 Route creation unit 4 Route selection unit 5 Score calculation unit 6 Similarity determination unit 7 Score adjustment unit 8 Optimal route selection unit 9 Local route determination unit 10 Score update unit 100 Route determination system 110 Position and attitude acquisition unit 120 Destination pose information 130 Control unit 140 Map information 150 Sensor 160 Plan creation unit 162 Global planning module 163 Global cost map 164 Global planning unit 165 Local planning module 166 Local cost map 167 Local planning unit 200 Route 205 Multiple intermediate points 210 Current position 215 Vehicle orientation at current position 220 Destination position 225 Desired orientation of vehicle at destination position 230 Intermediate point 1000, 2000 Route determination device 1001 CPU 1002 ROM 1003 RAM 1004 Bus 1005 Input / Output Interface 1006 Input Unit 1007 Output Unit 1008 Storage Unit 1009 Communication Unit 1010 Drive 1011 Magnetic Disk 1012 Optical Disk 1013 Flexible Disk 1014 Semiconductor Memory 3000 Computer DP Destination Pose Information GP Global Route LP Local Route

Claims

a route creation unit that creates a plurality of routes from a current position of a vehicle to the selected relay point; a route selection unit that selects a single route from the created plurality of routes without overlapping; a score calculation unit that calculates a score for evaluating the single selected route based at least on a length of the single route; a similarity determination unit that determines a similarity between the multiple routes and a previous route that is a route to the selected relay point that has been previously determined; a score adjustment unit that adjusts the calculated score if each of the multiple routes is similar to the previous route so that each of the multiple routes is evaluated higher; an optimal route selection unit that selects the route with the highest evaluation as an optimal route to the selected relay point based on the scores of each of the multiple optimal routes determined for the multiple relay points as a local route.

2. The route determination device of claim 1, wherein the route is created as a Reeds-Shepp curve, and the similarity determination unit determines that the previous route and each of the multiple routes are similar when a predetermined number of sequences at the beginning of the Reeds-Shepp curve that constitutes the previous route are identical to the predetermined number of sequences at the beginning of the Reeds-Shepp curve that constitutes each of the multiple routes.

3. The route determination device according to claim 1 or 2, wherein the score is the length of the one route, and the smaller the value of the score, the higher the evaluation of the one route.

4. The route determination device according to claim 1 or 2, wherein, when the previous route and each of the plurality of routes are similar, the score adjustment unit reduces the score calculated by the score calculation unit by a predetermined value.

5. A route determination device as described in claim 1 or 2, wherein the score calculation unit calculates a maximum value and an average value of the cost on the score map through which the one route passes based on a score map indicating a score indicating the effect of an obstacle on vehicle driving, and calculates the score by adding values ​​based on the calculated maximum value and average value of the cost to a value based on the distance of the one route.

6. The route determination device of claim 1 or 2, further comprising a score update unit that calculates maximum and average costs on the score map through which each of the multiple optimal routes determined for the multiple relay points passes based on a score map that indicates a score indicating the effect of an obstacle on vehicle travel, and updates the score by adding values ​​based on the calculated maximum and average costs to the score of each of the multiple optimal routes determined for the multiple relay points.

7. A route determination method comprising: obtaining a global route having a plurality of relay points; selecting one relay point from the plurality of relay points without overlapping; creating a plurality of routes from a current position of a vehicle to the selected relay point; selecting one route from the plurality of created routes without overlapping; calculating a score for evaluating the one selected route based at least on its length; determining a similarity between the plurality of routes and a previous route which is a route to the selected relay point that was previously determined; if each of the plurality of routes is similar to the previous route, adjusting the calculated score so that each of the plurality of routes is evaluated higher; selecting the route with the highest evaluation based on the score of each of the plurality of routes as an optimal route to the selected relay point; and determining the route with the highest evaluation as a local route based on the score of each of the plurality of optimal routes determined for the plurality of relay points.

8. A non-transitory computer-readable medium having stored thereon a program that causes a computer to execute the following processes: obtaining a global route having a plurality of relay points; selecting one relay point from the plurality of relay points without overlapping; creating a plurality of routes from a current position of a vehicle to the selected relay point; selecting one route from the plurality of created routes without overlapping; calculating a score for evaluating the one selected route based at least on the length of the route; determining a similarity between the plurality of routes and a previous route that is a route to the selected relay point that was previously determined; if each of the plurality of routes is similar to the previous route, adjusting the calculated score so that each of the plurality of routes is evaluated higher; selecting the route with the highest evaluation based on the score of each of the plurality of routes as an optimal route to the selected relay point; and determining the route with the highest evaluation as a local route based on the score of each of the plurality of optimal routes determined for the plurality of relay points.

Images