Movement route adaptive planning system of wheel-train robot

Abstract

The invention discloses an adaptive planning system for a motion route of a wheel-train robot, and relates to the technical field of route planning of wheel-train robots. A route planning module obtains work record data from a database, counts the type, position and occurrence time of an obstacle in a target area based on the work record data, draws an obstacle hotspot map, and plans an initial route in combination with task parameters. And finally, carrying out adaptive standby optimization on the initial route through the obstacle hotspot map to obtain a motion route. According to the technical scheme, the obstacle hotspot map is drawn through the working record data, a high-risk obstacle area can be recognized in advance, the calculation frequency of a real-time algorithm is reduced, and motion fluctuation caused by frequent path updating is avoided; by means of a standby route planned in advance, when the robot encounters an obstacle, real-time calculation is not needed, the route is directly switched, the obstacle avoidance response time is shortened, and the problem of path lag is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wheel train robot route planning, and specifically relates to a motion route adaptability planning system of a wheel train robot. BACKGROUND

[0002] Wheel train robots have a wide range of applications in multiple fields due to their simple structure, high motion efficiency, and relatively easy control. For example, they are used for cargo handling and sorting in industrial production and logistics, and for room cleaning in home services.

[0003] In a dynamic environment, the core of wheel train robot path planning lies in the contradiction between the uncertainty of environmental information and the real-time requirement. When the route planning algorithm cannot match the speed of dynamic changes in the environment, not only will the path lag, but also serious problems such as collision may be caused; for example, when a cleaning robot encounters a child during cleaning service, the route planning algorithm needs to generate an obstacle avoidance path within 0.5 seconds, but the child may enter the middle of the obstacle avoidance path within the 0.5 seconds. In addition, in order to respond to environmental changes, path planning needs to be iterated frequently, causing the motion state of the robot to fluctuate frequently, which will lead to unstable path updates and the inability to find a globally optimal path; for example, a wheel train robot in a factory encounters frequently moving shelves, and re-plans the path every 0.1 seconds, causing the wheel train robot to repeatedly turn and start and stop.

[0004] To solve the above technical problems, the application provides a motion route adaptability planning system of a wheel train robot. SUMMARY

[0005] The application aims to at least solve one of the technical problems existing in the prior art; for this purpose, the application provides a motion route adaptability planning system of a wheel train robot, which draws an obstacle hotspot map through work record data, can identify high-risk obstacle areas in advance, reduces the calculation frequency of real-time algorithms, and avoids motion fluctuations caused by frequent path updates; the pre-planned backup route allows the robot to switch routes directly without real-time calculation when encountering obstacles, shortens the obstacle avoidance response time, and solves the path lag problem.

[0006] To achieve the above purpose, the first aspect of the application provides a motion route adaptability planning system of a wheel train robot, which comprises a route planning module and a database connected thereto. The route planning module is used to obtain work record data from the database and draw an obstacle hotspot map of a target area based on the work record data; wherein the target area is the working area of the wheel train robot; and The basic route of the wheel train robot in the target area is planned based on task parameters; and the motion route of the wheel train robot is obtained by adaptively optimizing the basic route through the obstacle hotspot map.

[0007] In a possible implementation, the obstacle hot spot map of the target area is drawn based on the work record data, comprising: extracting the work record data, and counting obstacle records in the target area according to the work record data; wherein the obstacle records comprise obstacle types, positions where the obstacles are located, and obstacle occurrence times; drawing the obstacle hot spot map of the target area according to the obstacle records.

[0008] In a possible implementation, the obstacle hot spot map of the target area is drawn according to the obstacle records, comprising: drawing a framework map of the target area, and counting total times of occurrence of obstacles at each pixel point in the framework map according to the obstacle records; performing value assignment processing on each pixel point by comparing the total times with a preset number threshold to obtain a basic hot spot map; and adjusting the basic hot spot map in combination with robot parameters of the wheel train robot to obtain the obstacle hot spot map.

[0009] In a possible implementation, the value assignment processing on each pixel point by comparing the total times with a preset number threshold comprises: counting a number of the work record data, and obtaining a number threshold by multiplying the number by an adjustment coefficient; wherein the adjustment coefficient is less than 1; when the total times are greater than the number threshold, the corresponding pixel point is assigned a value of 1; otherwise, the corresponding pixel point is assigned a value of 0.

[0010] In a possible implementation, the basic hot spot map is adjusted in combination with the robot parameters of the wheel train robot, comprising: identifying obstacle regions in the basic hot spot map according to the value assignment processing result; judging whether spaces between adjacent obstacle regions are allowed to pass through the wheel train robot based on the robot parameters; if not, the adjacent obstacle regions are merged to obtain new obstacle regions.

[0011] In a possible implementation, the basic route is adaptively and redundantly optimized through the obstacle hot spot map, comprising: judging whether the basic route intersects with obstacle regions in the obstacle hot spot map; if so, the intersecting obstacle regions are marked as to-be-optimized regions, and intersection points are extracted; planning a motion route of the wheel train robot bypassing the to-be-optimized regions based on the intersection points as a redundant route one; and splicing and fusing the redundant route one with the basic route to obtain the motion route.

[0012] In a possible implementation, after the motion route of the wheel train robot bypassing the to-be-optimized regions is planned based on the intersection points, the method further comprises: verifying whether the distance between the adjacent to-be-optimized areas is less than a preset distance threshold; if yes, planning a movement route that bypasses the adjacent to-be-optimized areas simultaneously as a backup route two based on the intersection point; splicing and fusing the backup route one, the backup route two and the basic route to obtain the movement route.

[0013] In one possible implementation, verifying whether the distance between the adjacent to-be-optimized areas is less than a preset distance threshold comprises: marking the intersection points of the to-be-optimized areas and the basic route as feature point one and feature point two; wherein feature point one is closer to the initial position, and feature point two is closer to the target position; calculating the route distance between feature point two of the previous to-be-optimized area and feature point one of the next to-be-optimized area; when the route distance is less than a preset distance threshold, determining that the distance between the adjacent to-be-optimized areas is less than the preset distance threshold.

[0014] In one possible implementation, the total number of times that each pixel point in the framework map appears with an obstacle is counted according to the obstacle record, comprising: setting a statistical interval, determining a plurality of statistical time periods according to the statistical interval; associating the obstacle record with the plurality of statistical time periods according to the collection time; counting the total number of times that each pixel point in the framework map appears with an obstacle within the plurality of statistical time periods based on the associated obstacle record.

[0015] In one possible implementation, the total number of times within the plurality of statistical time periods and a preset number threshold are compared to perform value assignment processing on each pixel point to obtain a time period hotspot map. integrating all the time period hotspot maps according to the time sequence of the plurality of statistical time periods to obtain a basic hotspot map.

[0016] Compared with the prior art, the present application has the following advantages: 1. In the present application, the route planning module obtains work record data from a database, counts the type, position and appearance time of obstacles in the target area based on a plurality of work record data, draws an obstacle hotspot map, plans an initial route in combination with task parameters, and finally optimizes the initial route adaptively through the obstacle hotspot map to obtain a movement route. This technical solution draws an obstacle hotspot map through work record data, can identify high-risk obstacle areas in advance, reduces the calculation frequency of real-time algorithms, and avoids movement fluctuations caused by frequent path updates. The backup route planned in advance can allow the robot to switch routes directly without real-time calculation when encountering obstacles, shorten the obstacle avoidance response time, and solve the path lag problem.

[0017] 2. The application draws a hot spot map of obstacles, first counts the total number of obstacle occurrences of each pixel point in the target area framework map, sets a threshold value based on the amount of work record data to obtain a basic hot spot map, and then combines adjacent obstacle areas that cannot be passed according to the robot parameters; when optimizing the backup route, first plan the backup route one of a single obstacle area, then verify the distance of adjacent areas to be optimized, plan the backup route two that can bypass multiple close-range obstacle areas at the same time, and finally fuse the initial route with the two types of backup routes; the technical scheme can accurately distinguish between obstacles and passable areas through detailed obstacle hot spot map drawing, avoiding redundancy or lack of backup routes; the two types of backup routes planned in layers can not only deal with single obstacles, but also solve the frequent turning caused by multiple close-range obstacles, reducing motion adjustment, reducing energy consumption and hardware loss, and ensuring efficient and continuous task execution. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0019] Figure 1 The system principle diagram of the motion route adaptability planning system in embodiment one of the present application Figure 1 ; Figure 2 The system principle diagram of the motion route adaptability planning system in embodiment one of the present application Figure 2 ; Figure 3 The schematic diagram of the basic hot spot map in embodiment one of the present application Figure 4 The planning schematic diagram of the basic route in embodiment one of the present application Figure 5 The intersection schematic diagram of the basic route and the obstacle area in embodiment one of the present application Figure 6 The planning schematic diagram of the backup route one in the room cleaning scene in embodiment one of the present application Figure 7 The planning schematic diagram of the backup route one in the goods carrying scene in embodiment one of the present application Figure 8 The planning schematic diagram of the backup route two of adjacent areas to be optimized in embodiment one of the present application. DETAILED DESCRIPTION

[0020] The technical solutions of the present application will be described clearly and completely below in connection with the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0021] Embodiment one: Please refer to Figures 1-2 The first aspect of the present application provides a motion route adaptive planning system for a wheeled robot, which comprises a route planning module and a database connected thereto. The route planning module is configured to obtain work record data from the database, draw an obstacle hotspot map of a target area based on the work record data, and plan a basic route of the wheeled robot in the target area based on task parameters; and the basic route is adaptively and redundantly optimized based on the obstacle hotspot map to obtain a motion route of the wheeled robot.

[0022] The motion route adaptive planning system for the wheeled robot provided by the embodiments of the present application mainly comprises a route planning module and a database. The route planning module is configured to plan a motion route of the wheeled robot when performing a task, so as to avoid that the wheeled robot cannot perform efficient obstacle avoidance due to dynamic changes of a working environment. The database stores work record data obtained by the wheeled robot in a previous working process, which mainly includes working time and encountered obstacles.

[0023] Figure 1 The route planning module and the database can be independently set up from the wheeled robot, and are applied to an application scenario in which multiple wheeled robots work simultaneously, for example, multiple wheeled robots are simultaneously used for cargo handling in a factory. The route planning module and the database do not need to be integrated in the wheeled robot. The database stores work record data of multiple wheeled robots, and the route planning module draws or updates an obstacle hotspot map according to the work record data of the multiple wheeled robots. When a wheeled robot needs to perform a task, a motion route of the wheeled robot is planned according to its task parameters and the latest obstacle hotspot map, and the wheeled robot is controlled to perform a working task according to the motion route.

[0024] It should also be understood that the multiple wheeled robots in the application scenario only need to be able to complete a working task according to a motion route, and all the wheeled robots do not need to be able to collect work record data. If necessary, a special device can be set up to collect obstacle data in a target area, so as to be used for planning a motion route.

[0025] Figure 2The second system schematic diagram of the motion route adaptive planning system is that the route planning module and the database are integrated in the wheel system robot, and are applied to a scenario in which the wheel system robot works independently. For example, a cleaning service is provided by using a sweeping robot in household cleaning. The route planning module and the database are arranged on the wheel system robot. The wheel system robot collects working record data by using a sensor arranged on the wheel system robot, and sends the working record data to the database for storage. The route planning module draws or updates an obstacle hotspot map according to the working record data stored in the database, and plans a motion route of the wheel system robot for performing a working task according to the obstacle hotspot map.

[0026] Currently, a motion route planning scheme mainly combines a route planning algorithm built in the wheel system robot and collected environmental data. Since data processing and algorithm running need a certain time, the planned motion route may be lagged, and in a dynamically changing environment, the wheel system robot may run unstably due to frequent updating of the route. In view of this, the embodiment of the present application introduces an obstacle hotspot map, identifies a probability of encountering an obstacle by the wheel system robot in advance according to the obstacle hotspot map, and plans a standby route for bypassing the obstacle. During the execution of a task by the wheel system robot, the wheel system robot can immediately switch to the standby route as soon as an obstacle is encountered, and real-time planning by the route planning algorithm is not needed, so that route lag can be avoided.

[0027] The key to implementing the technical solution lies in drawing and updating the obstacle hotspot map. It is necessary to mine a moving rule of an obstacle in a target area by using sufficient working record data, and draw a probability of occurrence of the obstacle at each position in the target area, so as to draw the obstacle hotspot map.

[0028] In a preferred embodiment, the obstacle hotspot map of the target area is drawn based on the working record data, including: extracting the working record data, counting obstacle records in the target area according to the working record data, and drawing the obstacle hotspot map of the target area according to the obstacle records.

[0029] The working record data refers to video data or radar data collected by the wheel system robot in a target area during area scanning or execution of a task. An obstacle record of the target area can be extracted from the working record data, and the obstacle record includes an obstacle type, a position of the obstacle, an occurrence time of the obstacle, and the like. The obstacle type can be used to identify whether the obstacle belongs to a fixed obstacle or a mobile obstacle, so as to optimize the obstacle hotspot map. The position of the obstacle can be understood as an area covered by the obstacle, so as to be displayed in detail in the obstacle hotspot map, and facilitate subsequent planning of a standby route. The occurrence time of the obstacle refers to a time when the obstacle is identified in the working record data, and can also be used to optimize the obstacle hotspot map.

[0030] It should also be understood that the obstacle should be an obstacle that cannot be crossed by the wheel system robot, and the wheel system robot needs to re-plan a path to bypass the obstacle.

[0031] In the process of recording the obstacles in the target area according to the work record data, it is necessary to ensure that the work record data is sufficient and the coverage time reaches a certain time range. If the work record data is less or the coverage time range is short, the movement of most obstacles in the target area cannot be covered, which will lead to an imperfect obstacle hotspot map, and the reliability is insufficient when used for motion route planning or optimization.

[0032] For example, Figure 1 In the application scenario shown, the database can receive work record data of multiple wheeled robots at the same time, and the number of work record data can meet the requirements in a very short time. Figure 2 In the application scenario shown, only one or a few wheeled robots provide work record data, and it takes a longer time or performs more frequent tasks to reach the required number of work record data.

[0033] After the obstacle records are extracted, the framework range of the target area can be drawn first, and then the positions of the obstacles in the obstacle records, the appearance times of the obstacles, etc. are rendered to the framework range, and the obstacle hotspot map can be drawn. The work record data collected subsequently can be used to update the obstacle hotspot map to ensure that the obstacle hotspot map conforms to the latest actual situation of the target area.

[0034] When the obstacle hotspot map of the target area is obtained, the obstacles in the obstacle hotspot map need to be verified. On the one hand, this verification process is to determine the high-risk areas in the target area and to avoid them when planning the motion route. On the other hand, it is to merge adjacent high-risk areas in the target area to improve the efficiency and reliability of the planned backup route.

[0035] In a preferred embodiment, the obstacle hotspot map of the target area is drawn according to the obstacle records, including: drawing a framework map of the target area, and counting the total number of times each pixel point in the framework map appears with an obstacle according to the obstacle records; performing value assignment processing on each pixel point by comparing the total number with a preset number threshold to obtain a basic hotspot map; and adjusting the basic hotspot map in combination with the robot parameters of the wheeled robot to obtain the obstacle hotspot map.

[0036] In the process of obtaining the obstacle hotspot map, a framework map of the target area is first drawn using a graphics drawing software. The area in the framework map can be understood as the working area of the wheeled robot. The obstacles in the target area are identified from the obstacle records, and the total number of times each pixel point in the framework map appears with an obstacle is counted.

[0037] The total number of times that an obstacle exists in a pixel in the framework map can be understood as the probability that the obstacle exists at the location. If the probability is large, a backup route needs to be set for the obstacle at the location when planning a motion route; otherwise, a backup route can not be set, and when the wheeled robot encounters an obstacle at this location, an obstacle avoidance route is planned using a route planning algorithm. In this way, a backup route is planned for high-risk locations, and a real-time planning scheme is still used for low-risk locations, which can well balance the contradiction between frequent route planning and route lag.

[0038] When determining whether a backup route needs to be planned, the total number of times is compared with a preset number threshold to determine which pixels often have obstacles, and if they often have obstacles, they are determined to be high-risk pixels. The area composed of adjacent high-risk pixels is the obstacle area, and the pixels in the framework map can be divided into two categories through assignment processing, and a basic hotspot map can be obtained.

[0039] Considering that the space between adjacent obstacle areas in the basic hotspot map can not allow the wheeled robot to pass, there can be obstacles when planning a backup route for one of the obstacle areas. Therefore, the basic hotspot map needs to be adjusted in combination with the parameters of the wheeled robot, and an obstacle hotspot map can be obtained after adjustment. The robot parameters include size, turning radius, minimum turning angle, etc.

[0040] In a preferred embodiment, the assignment processing of each pixel by comparing the total number of times with the preset number threshold includes: counting the number of work record data, multiplying the number by an adjustment coefficient to obtain a number threshold; when the total number of times is greater than the number threshold, the corresponding pixel is assigned a value of 1; otherwise, the corresponding pixel is assigned a value of 0.

[0041] When assigning each pixel in the framework map, it can be implemented based on the probability that the corresponding pixel has an obstacle, so the probability threshold is a very important factor in the probability judgment process. The number threshold in this embodiment is set according to the number of work record data, which can be multiplied by 0.5 (default) or 0.7 (set according to the specific scene, if it is necessary to avoid real-time planning of the route as much as possible, it can be set to 0.1) as the number threshold. The total number of times that an obstacle exists in each pixel is identified according to the obstacle record, and if the total number of times is greater than the number threshold, it means that there is a high probability that an obstacle exists at the pixel, and the pixel is assigned a value of 1 in the framework map, otherwise the pixel is assigned a value of 0, which can obtain a basic hotspot map that clearly distinguishes the risk of obstacles.

[0042] If an obstacle is identified in each work record data, i.e. the total number of times is equal to the number of work record data, it can be considered that the obstacle is a fixed obstacle, and the total number of times is certainly greater than the preset number threshold.

[0043] Figure 3A schematic diagram of the basic heat map. Figure 3 Each grid cell represents a pixel in the target area's corresponding frame image. Black grid cells represent fixed obstacles, such as fixed shelves in a factory scene or lockers in a home environment. Gray grid cells represent non-fixed obstacles, which can also be understood as mobile obstacles, such as movable equipment in a factory scene or chairs in a home environment.

[0044] In a preferred embodiment, the basic heat map is adjusted in combination with the robot parameters of the wheeled robot, including: identifying the obstacle area in the basic heat map according to the assignment processing result; judging whether the space between adjacent obstacle areas allows the wheeled robot to pass based on the robot parameters; if not, merging the adjacent obstacle areas to obtain a new obstacle area.

[0045] If the wheeled robot cannot pass through the space between adjacent obstacle areas in the basic heat map, it is necessary to consider bypassing the adjacent obstacle areas when planning the motion route. Therefore, the adjacent obstacle areas can be merged into one obstacle area, which can improve the planning efficiency of subsequent motion route planning.

[0046] like Figure 3 As shown in the figure, assuming that the wheeled robot's path width is calculated to be at least greater than one grid width based on the robot parameters of the wheeled robot, then Figure 3 If there is only one grid space between obstacle area 1 and obstacle area 2, the wheeled robot cannot pass between obstacle area 1 and obstacle area 2. Therefore, obstacle area 1 and obstacle area 2 are merged to generate one obstacle area. The new obstacle area consists of obstacle area 1, obstacle area 2 and the grid between them.

[0047] When planning a basic route in the target area according to the current position of the wheeled robot and the target position, the mobile obstacles in the target area do not need to be considered when planning the basic route, but the boundary of the target area and the fixed obstacles therein need to be considered.

[0048] When planning a wheeled robot's motion path, the robot's mission parameters, including its initial and target positions, are first acquired. Next, a basic path within the target area is planned based on these parameters. This basic path is then adaptively optimized using an obstacle heat map to determine the robot's motion path.

[0049] As before, the obstacle regions in the obstacle hotspot map are divided into fixed obstacles and mobile obstacles. In terms of probability, fixed obstacles are certain to exist, while mobile obstacles are highly probable to exist. Therefore, when planning the motion route of the wheeled robot, fixed obstacles should be considered first, i.e., the wheeled robot should bypass the fixed obstacles, while it is not necessary to bypass the mobile obstacles.

[0050] Figure 4 A schematic diagram of the planning of the basic route is shown in FIG. 4, where position A is the initial position and position B is the target position. When planning the basic route, only the fixed obstacles are considered, i.e., the black grids, and the mobile obstacles, i.e., the gray grids, are not considered. Therefore, the basic route can be planned as the solid line from position A to position B. Assuming that there are no mobile obstacles in the current target area, the task is most efficient when the wheeled robot executes the task along the basic route. Figure 4 Figure 4 Figure 4

[0051] In a preferred embodiment, the basic route is adaptively and redundantly optimized by the obstacle hotspot map, including: judging whether the basic route intersects with the obstacle regions in the obstacle hotspot map; if yes, marking the intersecting obstacle regions as to-be-optimized regions and extracting the intersection points; planning a motion route of the wheeled robot bypassing the to-be-optimized regions as a redundant route one based on the intersection points; and splicing and fusing the redundant route one and the basic route to obtain the motion route.

[0052] However, if the basic route intersects with the obstacle regions, it means that the wheeled robot may be blocked by the mobile obstacles appearing when the wheeled robot executes the task along the basic route. If the wheeled robot plans an obstacle-avoiding route in real time when the mobile obstacles appear, it may cause route lag and unstable operation of the wheeled robot.

[0053] To solve this problem, a redundant route needs to be planned in advance to avoid route lag and unstable operation. Before planning the redundant route one, the obstacle regions that may need to be bypassed are determined, and then the redundant route one is planned according to the intersection points of the basic route and the obstacle regions.

[0054] Figure 5 A schematic diagram of the intersection of the basic route and the obstacle regions is shown in FIG. 5. It can be seen that there are intersection points C1 and C2 between the basic route and the obstacle regions. That is, there is a high probability that the wheeled robot will be blocked by the obstacle regions when the wheeled robot executes the task along the basic route. Therefore, the redundant route one needs to be planned.

[0055] Figure 6 ​​​The diagram below shows the planning of the backup route 1 in the room cleaning scenario. When the wheeled robot performs the cleaning task along the basic route, it may encounter obstacles. Considering its cleaning task, the backup route 1 planned based on the intersection points C1 and C2 is along the boundary of the obstacle area, that is, from the intersection point C1 to the right, and then upward to the intersection point C2, as shown in the figure. Figure 6 The solid arrow line between the intersection point C1 and the intersection point C2.

[0056] Figure 7 This is a schematic diagram of the planning of the backup route 1 in the cargo handling scenario. If the wheeled robot performs the cargo handling task along the basic route, it may encounter obstacles. Considering that the cargo handling task only needs to quickly transport the cargo to the destination, the backup route 1 planned based on the intersection points C1 and C2 can bypass the obstacle area. Figure 7 The solid arrow line except the basic route.

[0057] like Figure 5 and Figure 6 For example, the starting point and end point of backup route 1 can be an intersection point, such as a room cleaning scenario; or they can be non-intersection points, such as a cargo handling scenario, but the starting point and end point must both be on the basic route and should be as close to the intersection point as possible.

[0058] In a preferred embodiment, after planning the movement route of the wheeled robot to bypass the area to be optimized based on the intersection point, it also includes: verifying whether the distance between adjacent areas to be optimized is less than a preset distance threshold; if so, planning a movement route that bypasses the adjacent areas to be optimized at the same time based on the intersection point as a second backup route; splicing and fusing the first backup route, the second backup route and the basic route to obtain a movement route.

[0059] The base route may pass through multiple obstacle areas simultaneously, resulting in multiple areas to be optimized. Generally, a backup route 1 is planned for each area to bypass it. However, if two areas to be optimized are close together, executing the task along backup route 1 may result in lower efficiency. In this case, the adjacent areas to be optimized should be treated as a single area to be optimized, and a backup route 2 should be planned to promptly respond to changing circumstances and improve task execution efficiency.

[0060] It is worth noting that the movement route includes the basic route, backup route one and backup route two. When executing a task, priority is given to following the basic route. If an obstacle is identified in the basic route, backup route one or backup route two is selected for obstacle avoidance. If there is no corresponding backup route one or backup route two, the built-in route planning algorithm is used to plan in real time to avoid the obstacle, and the route is still merged into the basic route after the obstacle avoidance is completed.

[0061] In a preferred embodiment, the step of verifying whether the distance between the adjacent to-be-optimized areas is less than the preset distance threshold comprises: marking the intersection of the to-be-optimized area and the base route as a first feature point and a second feature point, wherein the first feature point is closer to the initial position and the second feature point is closer to the target position; calculating the route distance between the second feature point of the previous to-be-optimized area and the first feature point of the next to-be-optimized area; and determining that the distance between the adjacent to-be-optimized areas is less than the preset distance threshold when the route distance is less than the preset distance threshold.

[0062] Figure 8 a schematic diagram of planning the second alternate route for the adjacent to-be-optimized areas. Figure 8 In the case shown in FIG. 6, the base route intersects with two to-be-optimized areas at intersection points C1 and C2 and intersection points C3 and C4, respectively. The distance between the intersection points C2 and C3 on the base route is taken as the route distance, and if the route distance is less than the preset distance threshold, it is determined that the distance between the adjacent to-be-optimized areas is less than the preset distance threshold. If there are obstacles in both to-be-optimized areas when performing a task, the robot should avoid the obstacles according to the first alternate route for each to-be-optimized area, but after avoiding one to-be-optimized area along the first alternate route Y1, the robot needs to avoid the other to-be-optimized area along the second alternate route Y2 after advancing a certain distance along the base route. This process will reduce the efficiency of the wheel train robot and also affect the stability of the wheel train robot. In view of the above, the second alternate route Y3 is planned, which can directly avoid two to-be-optimized areas with a small distance, and the starting point of the second alternate route Y3 should be the starting point of the first alternate route Y1, and the ending point of the second alternate route Y3 should be the ending point of the first alternate route Y2.

[0063] It should be noted that if the distance between the adjacent to-be-optimized areas is small, the wheel train robot may need to immediately bypass the second to-be-optimized area along another first alternate route after bypassing one to-be-optimized area along the first alternate route. If the two to-be-optimized areas are close to each other, the motion efficiency of the wheel train robot will be reduced, and it is better to directly cross the two to-be-optimized areas, i.e., to plan the second alternate route. The preset distance threshold can be set according to the turning radius of the wheel train robot, for example, the preset distance threshold is set to be more than twice the turning radius.

[0064] It should be further noted that the fusion of the first alternate route, the second alternate route and the base route means that the first alternate route and the second alternate route are spliced on the base route to form a motion route with the base route as the main trunk and the first alternate route or the second alternate route as the branch. For example, Figure 8 In the case shown in FIG. 6, the corresponding motion route includes the base route from the initial position A to the target position B, the first alternate route Y1 bypassing the first to-be-optimized area, the second alternate route Y2 bypassing the second to-be-optimized area, and the second alternate route Y3 bypassing both to-be-optimized areas.

[0065] After the motion route planning is completed, the wheeled robot starts from the initial position A along the basic route, and if no obstacle affecting the passage is identified in the first to-be-optimized region, the wheeled robot still advances along the basic route, and if no obstacle affecting the passage is identified in the second to-be-optimized region, the wheeled robot advances along the basic route to the target position B. If an obstacle affecting the passage is identified in the first to-be-optimized region, the wheeled robot is controlled to advance along the standby route two Y3, and then join the basic route and reach the target position B, or advance along the standby route one Y1, and then join the basic route. If no obstacle affecting the passage is identified in the first to-be-optimized region, and an obstacle affecting the passage is identified in the second to-be-optimized region, the wheeled robot bypasses the to-be-optimized region through the standby route one Y2.

[0066] It should be understood that, for the case of "if an obstacle affecting the passage is identified in the first to-be-optimized region", if an obstacle affecting the passage is identified in the second to-be-optimized region on this basis, the standby route two Y3 is preferentially selected, and if no obstacle affecting the passage can be identified or is not identified in the second to-be-optimized region, the standby route one Y1 is preferentially selected.

[0067] Embodiment two: In a preferred embodiment, the total number of times that each pixel point in the obstacle record statistical framework diagram appears as an obstacle is counted according to the total number of times that each pixel point appears as an obstacle in the obstacle record statistical framework diagram, including: setting a statistical interval, determining a plurality of statistical time periods according to the statistical interval; associating the obstacle records with the plurality of statistical time periods according to the collection time; and counting the total number of times that each pixel point in the associated obstacle record statistical framework diagram appears as an obstacle within the plurality of statistical time periods.

[0068] In a preferred embodiment, each pixel point is assigned a value by comparing the total number of times within the plurality of statistical time periods with a preset number threshold, to obtain a time period hotspot diagram; and all time period hotspot diagrams are integrated in chronological order of the plurality of statistical time periods to obtain a basic hotspot diagram.

[0069] It is considered that the wheeled robot may perform a task in a specific time period, such as setting a cleaning robot to perform cleaning work at ten o'clock in the morning. If the motion route is planned according to the obstacle hotspot diagram drawn based on the work record data of all time periods, it is obviously not in line with the actual task requirements.

[0070] The obstacles appearing in the target area can be counted by time period, and the associated obstacle record in each statistical period is obtained. According to the obstacle record statistics, the total number of times of obstacles appearing in each pixel point in the frame diagram in the corresponding statistical period is counted, and the total number of times is assigned to the preset number threshold (set according to the number of working record data in the statistical period) to obtain the period hotspot diagram in each statistical period. The period hotspot diagrams of each statistical period are integrated in time sequence to obtain the basic hotspot diagram.

[0071] For example, assuming that the statistical interval is one hour, the day can be divided into twenty-four statistical periods, that is, the total number of times of obstacles appearing in each statistical period is obtained by counting the obstacle record in each hour. Then the total number of times is compared with the preset number threshold, and if it is greater than the preset number threshold, it is assigned to 1, otherwise it is assigned to 0, and the period hotspot diagram corresponding to the statistical period is obtained.

[0072] If the wheel series robot performs a task across the statistical period, considering the stability of the fixed obstacle, the basic route planned in the target area is fixed, that is, the probability or position of the moving obstacle may be different as time changes. For this case, the basic route is divided into each statistical period, such as dividing the basic route into statistical period one and statistical period two, that is, if the task is performed along the basic route, the first half is in statistical period one and the second half is in statistical period two. In the part of statistical period one, the standby route one is planned based on the period hotspot diagram corresponding to statistical period one, and in the part of statistical period two, the standby route one is planned based on the period hotspot diagram corresponding to statistical period two. In this case, standby route two can not be planned.

[0073] It is worth noting that if the wheel series robot cannot perform the task according to the motion route during the task execution process, or other unexpected situations occur, the built-in route planning algorithm is used to plan the route in real time to ensure the task execution. The route planning algorithm is, for example, A* algorithm, dynamic window method, etc.

[0074] The above embodiments are only used to illustrate the technical method of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A motion path adaptive planning system for a wheeled robot, characterized in that: The route planning module comprises a database connected thereto; The route planning module is configured to obtain work record data from the database, and draw an obstacle hotspot map of a target area based on the work record data, wherein the target area is a work area of the wheel train robot; and The wheel train robot is planned to have a basic route in the target area based on the task parameters; and the basic route is adaptively and redundantly optimized based on the obstacle hotspot map to obtain a motion route of the wheel train robot; wherein the adaptive and redundant optimization refers to planning a redundant route for the obstacle.

2. The motion path adaptive planning system of a wheel-legged robot according to claim 1, wherein, The obstacle hotspot map of the target area is drawn based on the work record data, and comprises: The work record data is extracted, and obstacle records in the target area are counted based on the work record data; wherein the obstacle records comprise an obstacle type, an obstacle position, and an obstacle occurrence time, and the obstacle is an obstacle that needs to be bypassed by the wheel train robot; The obstacle hotspot map of the target area is drawn based on the obstacle records.

3. The motion path adaptation planning system of a wheel-legged robot according to claim 2, wherein, The obstacle hotspot map of the target area is drawn based on the obstacle records, and comprises: A frame map of the target area is drawn, and a total number of times of occurrence of obstacles at each pixel point in the frame map is counted based on the obstacle records; Each pixel point is assigned based on a comparison between the total number of times and a preset number threshold to obtain a basic hotspot map; and the basic hotspot map is adjusted based on robot parameters of the wheel train robot to obtain the obstacle hotspot map.

4. The motion path adaptive planning system of a wheel-legged robot according to claim 3, wherein Each pixel point is assigned based on a comparison between the total number of times and a preset number threshold, and comprises: The number of the work record data is counted, and a number threshold is obtained by multiplying the number by an adjustment coefficient; wherein the adjustment coefficient is less than 1; If the total number of times is greater than the number threshold, the corresponding pixel point is assigned a value of 1; otherwise, the corresponding pixel point is assigned a value of 0.

5. The motion path adaptive planning system of a wheel-legged robot according to claim 3, wherein, The basic hotspot map is adjusted based on the robot parameters of the wheel train robot, and comprises: Obstacle regions in the basic hotspot map are identified based on the assignment result; It is judged whether a space between adjacent obstacle regions is allowed to be passed through by the wheel train robot based on the robot parameters; if not, the adjacent obstacle regions are merged to obtain a new obstacle region; wherein the robot parameters comprise a size, a turning radius, and a minimum turning angle.

6. The motion path adaptation system for a wheel-legged robot according to any one of claims 1 to 5, wherein The basic route is adaptively and redundantly optimized based on the obstacle hotspot map, and comprises: It is judged whether the basic route intersects with an obstacle region in the obstacle hotspot map; if so, the intersecting obstacle region is marked as a region to be optimized, and an intersection point is extracted; A motion route of the wheel train robot bypassing the region to be optimized is planned based on the intersection point as a redundant route one; and the redundant route one and the basic route are spliced and fused to obtain the motion route.

7. The motion path adaptation planning system of a wheel-legged robot according to claim 6, wherein, After the motion route of the wheel train robot bypassing the region to be optimized is planned based on the intersection point, the following steps are further included: It is verified whether a distance between adjacent regions to be optimized is less than a preset distance threshold; if so, a motion route bypassing the adjacent regions to be optimized is planned based on the intersection point as a redundant route two; wherein the preset distance threshold is set according to the turning radius of the wheel train robot; The redundant route one, the redundant route two, and the basic route are spliced and fused to obtain the motion route.

8. The motion path adaptation planning system of a wheel-legged robot according to claim 7, wherein, Verifying whether a distance between adjacent to-be-optimized areas is less than a preset distance threshold comprises: Marking an intersection of the to-be-optimized area and the basic route as a feature point one and a feature point two; the feature point one is closer to an initial position, and the feature point two is closer to a target position; Calculating a route distance between the feature point two of a previous to-be-optimized area and the feature point one of a next to-be-optimized area; when the route distance is less than a preset distance threshold, it is determined that the distance between adjacent to-be-optimized areas is less than the preset distance threshold.

9. The motion path adaptive planning system of a wheel-legged robot according to claim 3, wherein, According to the obstacle records, counting a total number of times of occurrence of obstacles for each pixel point in the framework map comprises: Setting a statistical interval, determining a plurality of statistical time periods according to the statistical interval; associating the obstacle records with the plurality of statistical time periods according to collection times; Based on the associated obstacle records, counting a total number of times of occurrence of obstacles for each pixel point in the framework map within the plurality of statistical time periods.

10. The motion path adaptation planning system of a wheeled robot according to claim 9, wherein, Through comparison of the total number of times within the plurality of statistical time periods and a preset number threshold, performing value assignment processing on each pixel point to obtain a time period heat map; Integrating all the time period heat maps according to a time sequence of the plurality of statistical time periods to obtain a basic heat map.

Images