Methods and apparatus for path planning of moving objects

Abstract

This invention relates to a method for path planning of a moving object. The method includes the following steps: generating multiple cost map layers (101) based at least on the positions of obstacles in a navigation area, the current positions and / or currently planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area; generating a master cost map (102) based on the multiple cost map layers; and planning a path to a target location based on the master cost map (103). This invention also relates to an apparatus for path planning of a moving object. Furthermore, this invention relates to a controller for a moving object, a moving object, and a computer-readable storage medium.

Description

Technical Field

[0001] This invention relates to a method and apparatus for path planning of a moving object. The invention also relates to a controller for the moving object. Furthermore, the invention relates to a moving object having the aforementioned apparatus or controller. Finally, the invention relates to a computer-readable storage medium. Background Technology

[0002] Currently, autonomous navigation path planning typically relies on data from onboard sensors (e.g., ultrasonic sensors, radar, cameras, lidar, etc.). A global path planner can plan the shortest global path based on a static map of the environment. Then, during navigation, if the onboard sensors detect obstacles in the initially planned path, the local path planner continuously adjusts the initially planned path. If multiple static or dynamic obstacles exist in the initially planned path, the global path planned by the global path planner may be far from optimal.

[0003] Currently, traffic flow can be improved by analyzing the location information of multiple moving objects in an environment based on ultra-wideband (UWB) systems. However, current solutions only consider the current location of the moving objects. Especially when navigating in larger environments, long routes must be planned. In this case, the current traffic conditions at a remote location may differ from the traffic conditions when the moving object arrives at that remote location. Summary of the Invention

[0004] The present invention aims to provide a method and an apparatus for path planning of moving objects. The method and apparatus for path planning of moving objects are primarily based on a hierarchical cost map and are capable of achieving an optimal navigation path to a target location.

[0005] Therefore, according to one aspect of the present invention, a method for path planning for a moving object or a method for autonomous navigation of a moving object is disclosed. The method includes: generating multiple cost map layers based at least on the positions of obstacles in a navigation area, the current positions and / or currently planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area; generating a main cost map based on the multiple cost map layers; and planning a path to a target location based on the main cost map.

[0006] The technical solution according to the present invention is based in particular on the following considerations. In the navigation area, there are not only static or dynamic obstacles but also multiple moving objects. For example, in an industrial area, in addition to static machines, there are often multiple moving objects, such as transport vehicles, robots, or people, etc., where these moving objects can be intelligent (e.g., artificial intelligence agents) or non-intelligent. In this case, when planning the optimal navigation path for a moving object to navigate from its starting position to its target position, it is necessary not only to avoid collisions with obstacles but also to avoid collisions with other moving objects. Furthermore, if other moving objects have path planning modules and communication modules, they can learn about their current planned paths and take this into account when planning their own paths. Additionally or alternatively (e.g., when a moving object does not have a path planning module or communication module), its historical position or historical movement pattern can be used to determine the path planning of the moving object.

[0007] According to one embodiment of the method, a first cost map layer is generated based on a static map of the navigation area. A second cost map layer is generated based on the current position of obstacles detected by the moving object during navigation. A third cost map layer is generated based on the current position and / or current planned path of the other moving objects. A fourth cost map layer is generated based on the historical positions of the other moving objects within a pre-given time period. Then, a main cost map is generated based on the first, second, third, and fourth cost map layers, or the first, second, third, and fourth cost map layers are integrated into a main cost map. Each cost map layer can be substantially a grid cell map. The cost values ​​of each cost map layer can be added or averaged (especially with assigned weight values) to produce the main cost map. The first, second, third, and fourth cost map layers (and possibly other cost map layers) are in the same cost coordinate system.

[0008] According to one embodiment of the method, the static map of the navigation area is pre-built.

[0009] According to one embodiment of the method, during navigation, the current position of obstacles in the navigation area is detected by one or more onboard sensors of the moving object. The onboard sensors can be selected from the group consisting of ultrasonic sensors, radar, lidar, cameras, or any other type of sensor. The moving object can dynamically detect / determine the position of obstacles in the navigation area using a single onboard sensor or a sensor network comprising multiple onboard sensors.

[0010] According to one embodiment of the method, the current position and / or planned path of other moving objects in a navigation area can be obtained based on a UWB system. The UWB system may, for example, include at least one tag attached to each moving object and at least four base stations located at different locations in the navigation area. Furthermore, the UWB system may also have a control unit connected to each base station, wherein the control unit is configured to calculate the moving object's position information using a Time of Flight (ToF) or Time Difference of Arrival (TDoA) method based on signals received from the tag at each anchor point. The calculated position information can be shared among moving objects via a UWB channel using the UWB protocol, allowing one moving object to know the current position of other moving objects. Additionally, if the moving object has a path planning module, its current planned path can be shared among moving objects via the UWB system.

[0011] According to one embodiment of the method, a density map is generated based on the historical positions of the other moving objects within a pre-given time period, and the fourth cost map layer is generated based on the density map. The density map specifically represents the historical movement patterns of the moving objects within the pre-given time period in the form of a density distribution. For a given grid cell, the more frequently the moving object passes through the grid cell within the pre-given time period, the higher the density value of that grid cell. In particular, grid cells with relatively high density values ​​are assigned relatively high cost values.

[0012] In particular, the start time and / or length of the pre-given time period can be adjusted in specific application scenarios.

[0013] According to one embodiment of the method, a (fourth) cost map layer can be generated not only based on the historical positions of other moving objects within a given time period, but also based on the historical positions of obstacles in the navigation area within a given time period. This is particularly advantageous in the case of dynamic obstacles. Furthermore, the fourth cost map layer can also be generated based on other information that may be useful for path planning.

[0014] According to one embodiment of the method, different cost map layers are generated based on different types of other moving objects. For example, when generating the fourth cost map layer, only the historical position of the robot or vehicle within a pre-given time period is considered. Additionally, a fifth cost map layer can be generated based on the historical position of a person within a pre-given time period, and a main cost map is generated accordingly based on the first, second, third, fourth, and fifth cost map layers. A similar situation applies to the third cost map layer.

[0015] According to one embodiment of the method, when generating the main cost map, weight values ​​can be assigned to the cost map layers, specifically different weight values. In specific application scenarios, the information of different cost map layers can be considered to varying degrees by adjusting the weight values ​​of each cost map layer.

[0016] According to one embodiment of the method, a higher weight value can be assigned to the third cost map layer, which represents the current position and / or current planned path of other moving objects, compared to the fourth cost map layer, which represents the historical position of other moving objects within a pre-given time period. This is highly advantageous when moving objects can share their current planned paths.

[0017] According to one embodiment of the method, a higher weight can be assigned to a fourth cost map layer, representing the historical positions of other moving objects over a pre-given time period, compared to a third cost map layer representing the current position and / or current planned path of other moving objects. This is highly advantageous when most other moving objects cannot share their current planned path.

[0018] According to one embodiment of the method, a higher weight value can be assigned to a fifth cost map layer representing the historical location of a person within a given time period, compared to a fourth cost map layer representing the historical location of a mobile robot or vehicle within a given time period. In this case, the person can be better protected.

[0019] Furthermore, according to another aspect of the present invention, an apparatus for path planning of a moving object or an apparatus for autonomous navigation of a moving object is disclosed. The apparatus includes at least a cost map layer generation module, a main cost map generation module, and a path planning module. The cost map layer generation module is configured to generate multiple cost map layers based at least on the positions of obstacles in a navigation area, the current positions and / or currently planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area. The main cost map generation module is configured to generate a main cost map based on the multiple cost map layers. The path planning module is configured to plan a path to a target location based on the main cost map.

[0020] According to one embodiment of the device, the cost map layer generation module can be further configured to: generate a first cost map layer based on a static map of the navigation area; generate a second cost map layer based on the current position of obstacles detected by moving objects during navigation; generate a third cost map layer based on the current position and / or current planned path of the other moving objects; and generate a fourth cost map layer based on the historical positions of the other moving objects within a pre-given time period. Correspondingly, the main cost map generation module can be further configured to generate a main cost map based on the first, second, third, and fourth cost map layers, or to integrate the first, second, third, and fourth cost map layers into a main cost map. The first, second, third, and fourth cost map layers (and possibly other cost map layers) are in the same cost coordinate system. Each cost map layer can be substantially a grid cell map. The cost values ​​of each cost map layer can be added or averaged to produce the main cost map.

[0021] According to one embodiment of the device, the static map of the navigation area is pre-built.

[0022] According to one embodiment of the device, during navigation, the current position of obstacles in the navigation area can be detected by one or more onboard sensors of the moving object. The onboard sensors can be selected from the group consisting of ultrasonic sensors, radar, lidar, cameras, or any other type of sensor. The moving object can dynamically detect / determine the position of obstacles in the navigation area by means of a single onboard sensor or a sensor network comprising multiple onboard sensors.

[0023] According to one embodiment of the device, the current position and / or planned path of other moving objects in a navigation area can be obtained based on a UWB system. The UWB system may, for example, include at least one tag attached to each moving object and at least four base stations located at different locations in the navigation area. Furthermore, the UWB system may also have a control unit connected to each base station, wherein the control unit is configured to calculate the moving object's position information using a Time of Flight (ToF) or Time Difference of Arrival (TDoA) method based on signals received from the tag at each anchor point. The calculated position information can be shared among moving objects via a UWB channel using the UWB protocol, allowing one moving object to know the current position of other moving objects. Additionally, if the moving object has a path planning module, its current planned path can be shared among the moving objects via the UWB system.

[0024] According to one embodiment of the apparatus, the cost map layer generation module can be further configured to: generate a density map based on the historical positions of the other moving objects within a pre-given time period, and generate the fourth cost map layer based on the density map. The density map specifically represents the historical movement patterns of the moving objects within the pre-given time period in the form of a density distribution. For a given grid cell, the more frequently the moving object passes through the grid cell within the pre-given time period, the higher the density value of the grid cell. In particular, grid cells with relatively high density values ​​can be assigned relatively high cost values.

[0025] In particular, the start time and / or length of the pre-given time period can be adjusted in specific application scenarios.

[0026] According to one embodiment of the device, the cost map layer generation module can be configured to generate a fourth cost map layer based not only on the historical positions of other moving objects within a predetermined time period but also on the historical positions of obstacles in the navigation area within a predetermined time period. This is particularly advantageous in the case of dynamic obstacles.

[0027] According to one embodiment of the device, the cost map layer generation module can be configured to generate different cost map layers based on different types of other moving objects. For example, the cost map layer generation module can be configured to generate a fourth cost map layer based on the historical position of a robot or vehicle within a pre-given time period. Additionally, the cost map layer generation module can be configured to generate a fifth cost map layer based on the historical position of a person within a pre-given time period. Correspondingly, the main cost map generation module can be configured to generate a main cost map based on the first cost map layer, the second cost map layer, the third cost map layer, the fourth cost map layer, and the fifth cost map layer. A similar approach applies to the third cost map layer.

[0028] According to one embodiment of the device, the main cost map generation module can be configured to assign different weights to the cost map layer when generating the main cost map.

[0029] According to one embodiment of the apparatus, the main cost map generation module can be configured to assign a higher weight value to the third cost map layer, which represents the current position and / or current planned path of other moving objects, compared to the fourth cost map layer, which represents the historical positions of other moving objects within a pre-given time period. This is highly advantageous when moving objects can share their current planned paths.

[0030] According to one embodiment of the device, the main cost map generation module can be configured to assign a higher weight value to a fourth cost map layer representing the historical positions of other moving objects over a pre-given time period, compared to a third cost map layer representing the current positions and / or planned paths of other moving objects. This would be highly advantageous when most of the other moving objects are people.

[0031] According to one embodiment of the device, the main cost map generation module can be configured to assign a higher weight value to a fifth cost map layer representing the historical location of a person within a given time period, compared to a fourth cost map layer representing the historical location of a mobile robot or vehicle within a given time period.

[0032] According to another aspect of the present invention, a controller for a moving object is disclosed. The controller may include: a processor; and a memory storing a computer program thereon, which, when executed by the processor, performs the steps of the method described above.

[0033] According to another aspect of the present invention, a moving object is disclosed. The moving object may include the path planning device for the moving object described above, and one or more vehicle-mounted sensors.

[0034] According to one embodiment of the moving object, the moving object may further include at least one UWB tag.

[0035] According to another aspect of the present invention, a computer-readable storage medium is disclosed. The computer-readable storage medium may have a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of the method described above. Attached Figure Description

[0036] Figure 1 A flowchart illustrating a method for path planning for a moving object or a method for autonomous navigation of a moving object according to an embodiment of the present invention is shown schematically.

[0037] Figure 2 A block diagram schematically illustrates a device for path planning of a moving object or a device for autonomous navigation of a moving object according to an embodiment of the present invention. Detailed Implementation

[0038] In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the claimed subject matter. However, the claimed subject matter can be practiced without these specific details. In some cases, methods, apparatus, or systems known to those skilled in the art have not been described in detail to highlight the claimed subject matter.

[0039] Figure 1 A schematic flowchart illustrating a method for path planning or autonomous navigation of a moving object is shown. The method is based on a cost map, particularly a hierarchical cost map, and is capable of achieving an optimal navigation path to a target location.

[0040] In method step 101, multiple cost map layers are generated based at least on the positions of obstacles in the navigation area, the current positions and / or current planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area.

[0041] Specifically, a first cost map layer is generated based on a static map of the navigation area. A second cost map layer is generated based on the current positions of obstacles detected by the moving object during navigation. A third cost map layer is generated based on the current positions and / or current planned paths of the other moving objects. A fourth cost map layer is generated based on the historical positions of the other moving objects within a pre-given time period. Then, a main cost map is generated based on the first, second, third, and fourth cost map layers, or the first, second, third, and fourth cost map layers are integrated into a main cost map.

[0042] Specifically, the first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer (and possibly other cost map layers) are in the same cost coordinate system. Each cost map layer can be essentially a grid cell map. Each grid cell of the cost map layer describes the probability that a corresponding coordinate location in the environment is occupied.

[0043] In particular, the static map of the navigation area is pre-built.

[0044] Specifically, during navigation, the current position of obstacles in the navigation area is detected by one or more onboard sensors of the moving object. The onboard sensors can be selected from the group consisting of ultrasonic sensors, radar, lidar, cameras, or any other type of sensor. The moving object can dynamically detect / determine the position of obstacles in the navigation area using a single onboard sensor or a sensor network comprising multiple onboard sensors.

[0045] Specifically, the current location and / or planned path of other moving objects in the navigation area can be obtained based on the UWB system. The UWB system may, for example, include at least one tag attached to each moving object and at least four base stations located at different locations in the navigation area. Furthermore, the UWB system may also have a control unit connected to each base station, wherein the control unit is configured to calculate the moving object's location information using a Time-of-Flight (ToF) method or a Time Difference of Arrival (TDoA) method based on signals received from the tag at each anchor point. The calculated location information can be shared among moving objects via a UWB channel using the UWB protocol, allowing one moving object to know the current location of other moving objects. Additionally, if the moving object has a path planning module, its current planned path can be shared among moving objects via the UWB system.

[0046] Specifically, a density map is generated based on the historical positions of the other moving objects within a pre-given time period, and the fourth cost map layer is generated based on the density map. The density map specifically represents the historical movement patterns of the moving objects within the pre-given time period in the form of a density distribution. For a given grid cell, the more frequently the moving object passes through the grid cell within the pre-given time period, the higher the density value of that grid cell. In particular, grid cells with relatively high density values ​​can be assigned relatively high cost values.

[0047] For example, this can be achieved by counting the number of times the grid cells have been sampled (e.g., sampling rate = 1 / s or higher) n. sampled (cell) represents the number n of other moving objects assigned to a given grid cell within a pre-defined time period (historical positions (and / or obstacle detections)).detected The density value d(cell) of the grid cell is calculated (e.g., within the range of [0.0, 1.0]) and is continuously updated in particular: d(cell) = n detected (cell) / n sampled (cell). To obtain a cost map layer, the information of the grid cells in the density map can be interpreted as the "cost" of a moving object moving through the corresponding grid cell. According to one implementation, the density map can be used directly as a cost map layer.

[0048] In particular, the start time and / or length of the pre-given time period can be adjusted in specific application scenarios.

[0049] Alternatively, a (fourth) cost map layer is generated based not only on the historical positions of other moving objects within a given time period, but also on the historical positions of obstacles in the navigation area within a given time period.

[0050] Alternatively, different cost map layers can be generated based on other moving objects of different types. For example, when generating the fourth cost map layer, only the historical position of the robot or vehicle within a pre-given time period is considered. Additionally, a fifth cost map layer can be generated based on the historical position of a person within a pre-given time period, and a main cost map can be generated accordingly based on the first, second, third, fourth, and fifth cost map layers. A similar approach applies to the third cost map layer.

[0051] In method step 102, a master cost map is generated based on the plurality of cost map layers.

[0052] Specifically, when generating the main cost map, weight values ​​can be assigned to cost map layers, with different weight values ​​for each layer. In specific application scenarios, the information from different cost map layers can be considered to varying degrees by adjusting the weight values ​​of each layer.

[0053] For example, a higher weight can be assigned to the third cost map layer, which represents the current position and / or planned path of other moving objects, compared to the fourth cost map layer, which represents the historical position of other moving objects within a given time period. This is particularly advantageous when moving objects can share their current planned paths.

[0054] For example, a higher weight can be assigned to a fourth cost map layer, representing the historical locations of other moving objects over a pre-given time period, compared to a third cost map layer that represents the current location and / or planned path of other moving objects. This would be particularly advantageous if most of the other moving objects were people.

[0055] For example, a higher weight value can be assigned to a fifth cost map layer that represents the historical location of a person within a given time period, compared to a fourth cost map layer that represents the historical location of a mobile robot or vehicle within a given time period.

[0056] In method step 103, a path to the target location is planned based on the main cost map.

[0057] Path planning can be implemented based on the master cost map using classic path planning methods.

[0058] Figure 2 A schematic block diagram is shown of an apparatus for path planning of a moving object or an apparatus for autonomous navigation of a moving object according to an embodiment of the present invention.

[0059] According to the present invention, the device 2 includes at least a cost map layer generation module 201, a main cost map generation module 202, and a path planning module 203.

[0060] The cost map layer generation module 201 is configured to generate multiple cost map layers based at least on the positions of obstacles in the navigation area, the current positions and / or current planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area.

[0061] The main cost map generation module 202 is configured to generate a main cost map based on the multiple cost map layers.

[0062] The path planning module 203 is configured to plan a path to the target location based on the main cost map.

[0063] Specifically, the cost map layer generation module 201 is further configured to: generate a first cost map layer based on a static map of the navigation area; generate a second cost map layer based on the current position of obstacles detected by the moving object during navigation; generate a third cost map layer based on the current position and / or current planned path of the other moving objects; and generate a fourth cost map layer based on the historical positions of the other moving objects within a pre-given time period. Correspondingly, the main cost map generation module 202 is further configured to: generate a main cost map based on the first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer, or integrate the first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer into a main cost map. The first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer (and possibly other cost map layers) are in the same cost coordinate system.

[0064] In particular, the static map of the navigation area is pre-built.

[0065] Specifically, during navigation, the current position of obstacles in the navigation area is detected by one or more onboard sensors of the moving object. The onboard sensors can be selected from the group consisting of ultrasonic sensors, radar, lidar, cameras, or any other type of sensor. The moving object can dynamically detect / determine the position of obstacles in the navigation area using a single onboard sensor or a sensor network comprising multiple onboard sensors.

[0066] Specifically, the current location and / or planned path of other moving objects in the navigation area can be obtained based on the UWB system. Location information can be shared between moving objects via UWB channels using the UWB protocol, allowing one moving object to know the current location of other moving objects. Furthermore, if a moving object has a path planning module, its current planned path can be shared among moving objects via the UWB system.

[0067] Specifically, the cost map layer generation module 201 can be further configured to: generate a density map based on the historical positions of the other moving objects within a pre-given time period, and generate the fourth cost map layer based on the density map. The density map, in particular, represents the historical movement patterns of the moving objects within the pre-given time period in the form of a density distribution.

[0068] In particular, the start time and / or length of the pre-given time period can be adjusted in specific application scenarios.

[0069] Specifically, the cost map layer generation module 201 is configured to generate a fourth cost map layer based not only on the historical positions of other moving objects within a given time period, but also on the historical positions of obstacles in the navigation area within a given time period.

[0070] Specifically, the cost map layer generation module 201 is configured to generate different cost map layers based on different types of other moving objects. For example, the cost map layer generation module 201 is configured to generate a fourth cost map layer based on the historical position of a robot or vehicle within a pre-given time period. Additionally, the cost map layer generation module 201 is configured to generate a fifth cost map layer based on the historical position of a person within a pre-given time period. Correspondingly, the main cost map generation module 202 is configured to generate a main cost map based on the first cost map layer, the second cost map layer, the third cost map layer, the fourth cost map layer, and the fifth cost map layer. A similar approach applies to the third cost map layer.

[0071] Specifically, the main cost map generation module 202 is configured to assign different weights to the cost map layer when generating the main cost map.

[0072] For example, the main cost map generation module 202 is configured to assign a higher weight value to the third cost map layer, which represents the current position and / or current planned path of other moving objects, compared to the fourth cost map layer, which represents the historical position of other moving objects within a pre-given time period.

[0073] For example, the main cost map generation module 202 is configured to assign a higher weight value to the fifth cost map layer, which represents the historical location of a person within a given time period, compared to the fourth cost map layer, which represents the historical location of a mobile robot or vehicle within a given time period.

[0074] While specific embodiments have been described, these embodiments are presented by way of example only and are not intended to limit the scope of the invention. The appended claims and their equivalents are intended to cover all modifications, substitutions, and alterations that fall within the scope and spirit of the invention.

Claims

1. A method for path planning of a moving object, the method comprising the following steps: Multiple cost map layers are generated based on at least the location of obstacles in the navigation area, the current location and / or current planned path of other moving objects in the navigation area, and the historical location of other moving objects in the navigation area. A main cost map is generated based on the multiple cost map layers; Based on the main cost map, a path to the target location is planned. Among them, a first cost map layer is generated based on the static map of the navigation area; A second cost map layer is generated based on the current position of obstacles detected by the moving object during navigation; A third cost map layer is generated based on the current position and / or current planned path of the other moving objects; A fourth cost map layer is generated based on the historical positions of the other moving objects within a pre-given time period; The main cost map is generated based on the first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer. Specifically, when generating the main cost map, weight values ​​are assigned to the plurality of cost map layers. The cost value of each cost map layer is summed or averaged using its assigned weight values ​​to generate the main cost map. Specifically, a density map is generated based on the historical positions of the other moving objects within a pre-given time period, and the fourth cost map layer is generated based on the density map. The density map represents the historical movement patterns of the other moving objects within the pre-given time period in the form of a density distribution.

2. The method according to claim 1, wherein, During navigation, the current position of the obstacle is detected by one or more onboard sensors of the moving object, wherein the one or more onboard sensors are selected from the group consisting of: Ultrasonic sensor; radar; LiDAR; Camera.

3. The method according to any one of claims 1 to 2, wherein, The current position and / or planned path of other moving objects are obtained based on the UWB system.

4. The method according to any one of claims 1 to 2, wherein, When the main cost map is generated, different weight values ​​are assigned to the cost map layer.

5. An apparatus for path planning of a moving object, the apparatus comprising the following modules: The cost map layer generation module is configured to generate multiple cost map layers based at least on the positions of obstacles in the navigation area, the current positions and / or current planned paths of other moving objects in the navigation area, and the historical positions of other moving objects in the navigation area. A main cost map layer generation module, configured to generate a main cost map based on the plurality of cost map layers; The path planning module is configured to plan a path to the target location based on the main cost map. in, The cost map layer generation module is further configured to: generate a first cost map layer based on a static map of the navigation area; generate a second cost map layer based on the current position of obstacles detected by the moving object during navigation; generate a third cost map layer based on the current position and / or current planned path of the other moving objects; and generate a fourth cost map layer based on the historical positions of the other moving objects within a pre-given time period. The main cost map generation module is further configured to generate the main cost map based on the first cost map layer, the second cost map layer, the third cost map layer, and the fourth cost map layer. The main cost map generation module is further configured to assign weight values ​​to the plurality of cost map layers when generating the main cost map. The cost value of each cost map layer is summed or averaged using its assigned weight values ​​to generate the main cost map. The cost map layer generation module is further configured to generate a density map based on the historical positions of the other moving objects within a pre-given time period, and to generate the fourth cost map layer based on the density map. The density map represents the historical movement patterns of the other moving objects within the pre-given time period in the form of a density distribution.

6. The apparatus according to claim 5, wherein, During navigation, the current position of the obstacle is detected by one or more onboard sensors of the moving object, wherein the one or more onboard sensors are selected from the group consisting of: Ultrasonic sensor; radar; LiDAR; Camera.

7. The apparatus according to any one of claims 5 to 6, wherein, The current position and / or current planned path of the other moving objects are obtained based on the UWB system.

8. The apparatus according to any one of claims 5 to 6, wherein, The main cost map generation module is also configured to assign different weight values ​​to the cost map layer when generating the main cost map.

9. A controller for a moving object, the controller comprising: processor; A memory having a computer program stored thereon, which, when executed by a processor, performs the steps of the method according to any one of claims 1-4.

10. A moving object, the moving object comprising: The apparatus for path planning of a moving object according to any one of claims 5 to 8 or the controller according to claim 9; One or more vehicle-mounted sensors.

11. The moving object according to claim 10, wherein, The moving object also includes at least one UWB tag.

12. A computer-readable storage medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it performs the steps of the method according to any one of claims 1 to 4.

Images