A Robot Global Path Planning Method and Apparatus Based on Road Network Topology
By calculating the access costs of the starting and ending points in the road network topology and combining them with a graph search algorithm, the problem of path planning deviation in the existing technology is solved, an accurate global path is generated, and the accuracy of path length estimation is improved.
Patent Information
- Application Number
- CN202611626904.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-20
- Publication Date
- 2026-11-17
AI Technical Summary
Existing global path planning schemes fail to effectively consider the cost of connecting the robot's starting and ending points to the road network topology, resulting in deviations between the planned path and the actual passable roads, and inaccurate path length estimations.
In path planning, the initial cost and tail cost are calculated by determining the node positions of the robot's start and end points in the road network topology, and the path search is performed by combining Dijkstra's or A* algorithm, incorporating access costs to generate an accurate global path.
It incorporates access costs into the road network topology, generates accurate global paths, reduces deviations in path planning, and improves the accuracy of path length estimation.
Smart Images

Figure CN123103715A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of automatic navigation technology, and in particular to a method and apparatus for global path planning of robots based on road network topology. Background Technology
[0002] Global path planning is the core component for robots to achieve autonomous navigation, and the quality of the planning results directly affects whether the robot can safely and efficiently reach its target location.
[0003] Existing global path planning schemes typically construct road network topology based on two-dimensional grid cost maps, and use graph search algorithms such as A* and Dijkstra to perform path search and output a global path along the road.
[0004] However, in practical applications, the starting and ending points of a planning request often do not correspond precisely to nodes in the road network topology. The starting and ending points must be connected to the road network topology before path searching can proceed, resulting in access costs between the starting and ending points and their respective access locations. Existing technologies such as CN120952496B directly perform path planning based on the access locations of the starting and ending points in the road network topology, without incorporating access costs into the path cost calculation. This leads to discrepancies between the final output global path and the actual passable roads, and inaccurate path length estimations. Therefore, how to consider access costs in path planning to obtain an accurate global path is a pressing technical problem that needs to be solved in this field. Summary of the Invention
[0005] In view of this, this disclosure proposes a robot global path planning method and apparatus based on road network topology, which can consider access costs in path planning to obtain an accurate global path.
[0006] According to a first aspect of this disclosure, a robot global path planning method based on road network topology is provided, comprising: Determine the region where the current path to be planned is located, obtain the corresponding road network file based on the region, and parse the road network file to generate the road network topology corresponding to the region; The starting point and ending point of the current planned path are determined from the received planning request, and the node position of the starting point and the node position of the ending point in the road network topology are determined. The initial cost is determined based on the starting point and the node position of the starting point in the road network topology, and the tail segment cost is determined based on the ending point and the node position of the ending point in the road network topology; Based on the location of the starting point in the road network topology, the location of the ending point in the road network topology, the initial cost, and the tail cost, path planning is performed, and the result of the path planning is used as the final planned path.
[0007] In one possible implementation, parsing the road network file to generate the road network topology corresponding to the region includes: Nodes and road edges are parsed from the road network file; A basic adjacency table is constructed based on the adjacency relationships between each node and each road edge, and the road network topology is constructed based on the information recorded in the basic adjacency table.
[0008] In one possible implementation, determining the node position of the starting point and the node position of the ending point in the road network topology includes: Determine the starting point access location conditions and the ending point location constraints based on the starting point and the ending point, respectively. Each node and each roadside that satisfies the starting point access location condition and the ending point location constraint condition is respectively identified as a candidate node and candidate roadside for the starting point and a candidate node and candidate roadside for the ending point. The node position of the starting point in the road network topology is determined based on the candidate nodes and candidate roadsides of the starting point, and the node position of the ending point in the road network topology is determined based on the candidate nodes and candidate roadsides of the ending point.
[0009] In one possible implementation, determining the node position of the starting point in the road network topology based on candidate nodes and candidate road edges of the starting point, and determining the node position of the ending point in the road network topology based on candidate nodes and candidate road edges of the ending point, includes: The node position of the starting point in the road network topology is determined based on the distance between the starting point and each candidate node of the starting point, and the distance between the starting point and each candidate roadside of the starting point. The node position of the endpoint in the road network topology is determined based on the distance between the endpoint and each candidate node of the endpoint, and the distance between the endpoint and each candidate roadside of the endpoint.
[0010] In one possible implementation, determining the initial cost based on the starting point and its node position in the road network topology, and determining the tail segment cost based on the ending point and its node position in the road network topology, includes: The distance between the starting point and the node position of the starting point in the road network topology is calculated as the initial cost; The distance between the endpoint and the node location of the endpoint in the road network topology is calculated as the tail segment cost.
[0011] In one possible implementation, the path planning based on the location of the starting point in the road network topology, the location of the ending point in the road network topology, and the initial cost and the tail cost includes: If the node position of the starting point and the node position of the ending point in the road network topology are located on the same road edge, then the road edge is directly used as the result of the path planning. If the node positions of the starting point and the ending point in the road network topology are not on the same road, then the source point and sink point are determined, and path planning that includes both the initial cost and the tail cost is executed, wherein the source point refers to the starting point for executing the path planning, and the sink point refers to the ending point for executing the path planning.
[0012] According to a second aspect of this disclosure, a robot global path planning device based on road network topology is provided, comprising: Road network file loading module: Determines the region where the current planned path is located, and obtains the corresponding road network file based on the region; Road network topology construction module: Receives planning request, determines the starting point and ending point of the route planning from the planning request, constructs road network topology based on the road network file, connects the starting point and the ending point to the road network topology, and determines the starting point connection position and the ending point connection position; Access cost calculation module: determines the initial cost based on the starting point and the node position of the starting point in the road network topology, and determines the tail segment cost based on the ending point and the node position of the ending point in the road network topology; Global path search module: Performs shortest path search based on the starting access location, the ending access location, the initial cost, and the tail segment cost, and uses the result of the shortest path search as the global path obtained by path planning.
[0013] In one possible implementation, the road network topology construction module parses nodes and road edges from the road network file, constructs a basic adjacency table based on the adjacency relationships between each node and each road edge, and constructs the road network topology based on the information recorded in the basic adjacency table.
[0014] According to a third aspect of this disclosure, a robot global path planning device based on road network topology is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the method described in the first aspect of this disclosure.
[0015] According to a fourth aspect of this disclosure, a non-volatile computer-readable storage medium is provided that stores computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the method described in the first aspect of this disclosure.
[0016] This disclosure provides a method and apparatus for robot global path planning based on road network topology. The method includes: determining the region where the current path to be planned is located, obtaining the corresponding road network file based on the region, and parsing the road network file to generate the road network topology of the region; determining the start point and end point of the current path to be planned from a received planning request, and determining the node positions of the start point and the end point in the road network topology; determining the initial cost based on the start point and the node positions of the start point and the end point in the road network topology, and determining the tail cost based on the end point and the node positions of the end point and the end point in the road network topology; and performing path planning based on the positions of the start point and the end point in the road network topology, as well as the initial cost and the tail cost, to obtain a globally planned path. By considering the initial cost and the tail cost in the path planning, this disclosure can obtain an accurate globally planned path.
[0017] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description
[0018] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.
[0019] Figure 1 A flowchart illustrating robot global path planning based on road network topology according to an embodiment of the present disclosure is shown; Figure 2 A system framework diagram of robot global path planning based on road network topology according to an embodiment of the present disclosure is shown. Figure 3 A flowchart illustrating electronic fence conflict detection for robot global path planning based on road network topology according to an embodiment of the present disclosure is shown. Figure 4 A flowchart illustrating a road network topology loading process for robot global path planning based on road network topology according to an embodiment of the present disclosure is shown. Figure 5 A flowchart illustrating the process of publishing the visualization results of a robot global path planning based on road network topology according to an embodiment of the present disclosure is shown. Figure 6 A flowchart illustrating a robot global path planning based on road network topology according to an embodiment of the present disclosure is shown. Figure 7A schematic block diagram of a robot global path planning device based on road network topology according to an embodiment of the present disclosure is shown. Figure 8 A schematic block diagram of a robot global path planning device based on road network topology according to an embodiment of the present disclosure is shown. Detailed Implementation
[0020] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.
[0021] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.
[0022] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.
[0023] <Method Implementation> Figure 1 A flowchart illustrating a robot global path planning method based on road network topology according to an embodiment of this disclosure is shown. Figure 1 As shown, the method includes steps S1100-S1400: S1100: Determine the area where the current planned path is located, obtain the corresponding road network file based on the area, and parse the road network file to generate the road network topology corresponding to the area.
[0024] First, it is understandable that road network files are digital files describing traffic information such as roads and intersections, and road network files are usually stored in regions. Therefore, when planning a route, it is necessary to first determine the region where the route to be planned is located, and then obtain the corresponding road network file based on the region.
[0025] In this disclosure, the road network documents adopt the following format: Figure 2The GeoJSON file implementation shown in the image records nodes describing intersection information and road edges describing road information. Nodes in the GeoJSON file are of type Point, representing the location of intersections in the road network and carrying node attributes such as node identifier and coordinates. Road edges in the GeoJSON file are of type MultiLineString, where a road edge consists of multiple polylines (i.e., a sequence of vertices of multiple connected polylines represents a road edge), carrying edge identifiers, start node identifiers, end node identifiers, and road edge attributes such as road direction, road length, road class, passable width, floor level, and map number.
[0026] It should be noted that since road edges in the road network file are composed of multiple polyline vertices connected sequentially, the distance between any two locations within a road edge is calculated by cumulatively adding the polyline segments between those two locations, rather than directly calculating the straight-line distance between them. For example, on a road edge... middle, arrive The distance between them includes arrive distance, arrive The distance.
[0027] In one possible implementation, such as Figure 3 As shown, the road network file may also include electronic fences, which are used to indicate prohibited areas. These electronic fences can be at least one of point fences, line segment fences, and polygon fences. Since roads covered by electronic fences are actually impassable, before constructing the road network topology, it is necessary to check in the road network file whether there are conflicts between each road edge and the electronic fence, and to mask any conflicting road edges. Specifically, for point fences, the prohibited area is a circular area with a conflict radius. The distance from the point fence to the road edge is determined to be less than the conflict radius; if it is, a conflict with the electronic fence exists. For line segment fences, it is determined whether the road edge intersects with the line segment fence; if they do, a conflict with the electronic fence exists. For polygon fences, it is determined whether the road edge is inside the polygon fence or intersects with it; if it is inside or intersects with the polygon fence, a conflict with the electronic fence exists. If a roadside conflicts with an electronic fence, it means that the roadside is actually impassable. This roadside will be hidden in the road network file and will not appear in the road network topology, nor will it be used in subsequent path planning. This ensures that all global paths obtained from the final path planning are actually passable paths.
[0028] Regarding the acquisition of road network files, in one possible implementation, the planning request carries the location information of the starting point and the location information of the ending point. The location information of the starting point and the location information of the ending point can be used as the location information of the path to be planned to determine the target area, and the corresponding road network file can be matched and loaded based on the target area. The target area is the area where the current path to be planned is located.
[0029] Specifically, in this disclosure, the planner performs global path planning for the robot. The planner refers to the system that executes the path planning method described above. Since road network files are stored by region, and the road network files for the same region iterate as the actual road conditions change, to avoid repeatedly loading road network files each time a planning request is received, and to avoid using expired road network files, after determining the target region, the planner first checks whether the road network file corresponding to the target region has been loaded in the planner and whether it needs to be updated: if the road network file corresponding to the target region has been loaded in the planner and does not need to be updated, then the road network topology is directly constructed using the road network file already loaded in the planner, without repeated loading; if the road network file corresponding to the target region has not yet been loaded in the planner, or although it has been loaded but needs to be updated due to version changes, then a method is called... Figure 2 and Figure 4 The “update_graph” service shown in the figure accesses the database that stores road network files, reads the road network file corresponding to the target area from the database, and loads or updates the road network file into the planner so that the planner can build the road network topology based on the road network file.
[0030] In one possible implementation, it can be determined whether a road network file needs updating by comparing its version identifier in the planner with the version identifier of the corresponding road network file in the database: if they match, no update is needed; if they don't match, an update is required. The version identifier can be set to either the road network file's version number or an update timestamp.
[0031] Furthermore, such as Figure 4 The content shown in the figure is obtained by parsing the nodes and road edges from the road network file, constructing a basic adjacency list based on the connection relationship between each node and each road edge, and building the road network topology.
[0032] Specifically, first, data of type "Point" is parsed from the road network file. This data is used as nodes to create a node table, recording node attributes such as node identifier and coordinates. Then, data of type "MultiLineString" is parsed from the road network file. This data is used to create a road edge table, recording road edge attributes such as edge identifier, start node identifier, end node identifier, road direction, road length, road class, passable width, floor level, and map number. It is important to note that when creating the road edge table, if the start or end node identifier of a road edge does not exist in the node table, that road edge is skipped. Further, a basic adjacency table is created based on the node table and the road edge table. Specifically, adjacency relationships are written in the basic adjacency table from the start node identifier to the end node identifier. When the road direction of the road edge is "bidirectional", the reverse adjacency relationship from the end node identifier to the start node identifier is written in the adjacency table. Finally, the road network topology is constructed based on the node attribute information recorded in the node table, the road edge attribute information recorded in the road edge table, and the adjacency relationships between each node and each road edge recorded in the basic adjacency table.
[0033] In one possible implementation, after constructing the road network topology, it can be visualized. Specifically, the visualization can be generated as follows: First, nodes are displayed as spheres, and different colors, line widths, or arrows are used to distinguish between one-way and two-way road edges. Node attribute information and road edge attribute information are displayed using text annotations. Then, a persistent QoS (Quality of Service) publishing mechanism is used... Figure 5 The persistent publishing of "RodeNetWork" shown above publishes the above visualization results to robot terminals in the vicinity that have path planning needs, so that relevant personnel can intuitively view the traffic conditions in the target area through the visualization interface on the robot terminal.
[0034] S1200 determines the start and end points of the current planned path from the received planning request, and determines the node position of the start point in the road network topology and the node position of the end point in the road network topology.
[0035] In real-world environments, the starting and ending points of a planned path often do not perfectly match the nodes in the road network topology. Therefore, it is necessary to match the node position of the starting point in the road network topology as the starting point access position and the node position of the ending point in the road network topology as the ending point access position. Only after determining the starting and ending point access positions can the road network topology be used for path planning.
[0036] In one possible implementation, firstly, the starting point location constraints and the ending point location constraints need to be determined based on the starting point and the ending point, respectively; then, each node and each roadside that satisfies the starting point location constraints and the ending point location constraints are respectively regarded as candidate nodes and candidate roadsides of the starting point and candidate nodes and candidate roadsides of the ending point; finally, the node position of the starting point in the road network topology is determined based on the candidate nodes and candidate roadsides of the starting point, and the node position of the ending point in the road network topology is determined based on the candidate nodes and candidate roadsides of the ending point.
[0037] Specifically, the coordinates of the starting point and the ending point are extracted from the planning request, and the starting point location constraints and ending point location constraints are determined based on the location information of the starting point and the ending point, respectively. The starting point location constraints are used to constrain the location conditions that nodes and road edges that can match the starting point must satisfy (i.e., the conditions that the attribute information of nodes and road edges must meet), while the ending point location constraints are used to constrain the location conditions that nodes and road edges that can match the ending point must satisfy.
[0038] In one possible implementation, the start-point and end-point constraints may include at least one of floor constraints and height constraints. Floor constraints constrain the relationship between the floor to which a node or roadside belongs and the floor to which the start or end point belongs. Height constraints constrain the relationship between the height of a node or roadside and the height of the start or end point. It should be noted that since the attribute information of roadsides in this disclosure includes floors, the road network topology may include nodes and roadsides belonging to multiple floors. Therefore, this disclosure may require cross-floor path planning; hence, the aforementioned floor constraints or height constraints are set as start-point and end-point constraints. Specifically, based on the position information of the start point in the vertical direction (i.e., the Z-axis direction in the three-dimensional coordinate system) in the planning request, the floor to which the start point belongs and the start-point height can be determined, thereby determining the start-point constraints. Based on the position information of the end point in the vertical direction, the floor to which the end point belongs and the end-point height can be determined, thereby determining the end-point constraints. The start-point and end-point constraints are independent of each other; the floors they constrain may be the same or different. After obtaining the starting point location constraints and the ending point location constraints, the floor and height of each node can be determined by the coordinate information in the node attribute information, and the floor of each roadside can be determined by the floor information in the roadside attribute information. This allows us to determine whether each node and each roadside meets the above constraints. Nodes and roadsides that meet the starting point location constraints are selected as candidate nodes and candidate roadsides for the starting point, respectively. Nodes and roadsides that meet the ending point location constraints are selected as candidate nodes and candidate roadsides for the ending point, respectively.
[0039] In addition, as an optional implementation method for constructing road network topology, such as Figure 4 The KD-tree construction shown in this disclosure allows for the storage and indexing of nodes in the road network topology using a KD-tree data structure, which can accelerate the retrieval of candidate nodes. Specifically, a KD-tree index is constructed using the coordinate information of each node as the key; nodes that satisfy the start-point and end-point constraints are selected as candidate nodes through the KD-tree index. Those skilled in the art will understand that, compared to linearly traversing all nodes to select candidate nodes, using a KD-tree index can significantly reduce time consumption, thereby improving overall response efficiency.
[0040] Furthermore, the node position of the starting point in the road network topology is determined based on the distance between each candidate node of the starting point and the distance between each candidate roadside of the starting point; the node position of the ending point in the road network topology is determined based on the distance between each candidate node of the ending point and the distance between each candidate roadside of the ending point.
[0041] In one possible implementation, the distances between the starting point and each candidate node are calculated using the location information of the starting point and the coordinate information of each candidate node. The candidate node closest to the starting point is selected as the nearest node. The starting point is then projected onto each candidate roadside to obtain projection points. The distances between the starting point and each candidate roadside are calculated using the location information of the starting point and the coordinate information of each projection point. The projection point closest to the starting point is selected as the nearest projection point. The node position of the starting point in the road network topology is then selected from the nearest node and the nearest projection point. Similarly, the distances between the ending point and each candidate node are calculated using the location information of the ending point and the coordinate information of each candidate node. The candidate node closest to the ending point is selected as the nearest node. The ending point is then projected onto each candidate roadside to obtain projection points. The distances between the ending point and each candidate roadside are calculated using the location information of the ending point and the coordinate information of each projection point. The projection point closest to the ending point is selected as the nearest projection point. The node position of the ending point in the road network topology is then selected from the nearest node and the nearest projection point.
[0042] It's also important to note that when projecting the start and end points of the planning request onto each candidate road edge, the projection arc length ratio is calculated. The projection arc length ratio refers to the ratio of the distance from each projection point to the starting node of the road edge to the length of the road edge. For example, if the vertex sequence of a certain road edge is... in As the starting node, the coordinates of the projection point on the side of this road are: Then the distance from the projection to the starting node of the roadside is The length of the road along the side of the road is The calculated projection arc length ratio is approximately .
[0043] S1300 determines the initial cost based on the starting point and the node position of the starting point in the road network topology, and determines the tail segment cost based on the ending point and the node position of the ending point in the road network topology.
[0044] In one possible implementation, determining the initial cost based on the starting point and its node position in the road network topology includes three methods: (1) If the node position of the starting point in the road network topology is determined by the nearest node, then the distance between the starting point and the nearest node is directly used as the initial cost; (2) If the node position of the starting point in the road network topology is determined by the nearest projection point, and the road direction of the road is "one-way", then the starting node of the road is taken as the source point. The source point refers to the starting point of the path planning in the road network topology. Then the distance between the starting point and the nearest projection point and the distance between the nearest projection point and the source point are calculated. The sum of these two distances is used as the initial cost. (3) If the node position of the starting point in the road network topology is determined by the nearest projection point, and the road direction of the road is "bidirectional", then the starting node and the ending node of the road are taken as the source points respectively. Then the distance between the starting point and the nearest projection point, and the distance between the nearest projection point and each source point are calculated. The sum of these two distances is used as the initial cost. It can be understood here that when the road direction of the road is "bidirectional", since there are two source points (i.e., Figure 6 The first source point and the second source point are shown in the figure, so the calculated initial cost also corresponds to two points.
[0045] Similarly, when determining the tail segment cost based on the endpoint and its node position in the road network topology, there are also three methods: (1) If the node position of the endpoint in the road network topology is determined by the nearest node, then the distance between the endpoint and the nearest node is directly used as the tail segment cost. (2) If the node position of the endpoint in the road network topology is determined by the nearest projection point, and the road direction of the road is "one-way", then the terminal node of the road is taken as the sink point. The sink point refers to the terminal point of the path planning in the road network topology. Then the distance between the endpoint and the nearest projection point and the distance between the nearest projection point and the sink point are calculated. The sum of these two distances is taken as the tail segment cost. (3) If the node position of the endpoint in the road network topology is determined by the nearest projection point, and the road direction of the road is "bidirectional", then the starting node and the ending node of the road are taken as sinks respectively. Then, the distance from the endpoint to the nearest projection point and the distance from the nearest projection point to each sink are calculated, and the sum of these two distances is taken as the tail segment cost. It can be understood here that when the road direction of the road is "bidirectional", since there are two sinks (i.e., Figure 6 The first candidate sink and the second candidate sink are shown in the figure, so there are also two corresponding tail segment costs calculated.
[0046] S1400 performs path planning based on the location of the starting point in the road network topology, the location of the ending point in the road network topology, the initial cost, and the tail cost, and uses the result of the path planning as the final planned path.
[0047] In one possible implementation, if the starting point and the ending point are located on the same road edge in the road network topology, the path planning result is directly determined based on that road edge.
[0048] Specifically, the first step is to determine whether the starting point can directly reach the ending point via the roadside. If the ratio of the projected arc length of the starting point to the ratio of the projected arc length of the ending point is less than or equal to the ratio of the projected arc length of the ending point, it means the starting point can directly reach the ending point via the roadside. If the ratio of the projected arc length of the starting point to the ratio of the projected arc length of the ending point is greater than the ratio of the projected arc length of the ending point, and the road direction of the roadside is "one-way," it means the starting point cannot directly reach the ending point via the roadside. If the ratio of the projected arc length of the starting point to the ratio of the projected arc length of the ending point is greater than the ratio of the projected arc length of the ending point, and the road direction of the roadside is "two-way," it means the starting point can reach the ending point via the roadside in the opposite direction. Then, when the starting point can reach the ending point via the roadside, the roadside is trimmed in the order from the starting point's projection point to the ending point's projection point. The trimmed path is used as the result of path planning. Trimming the roadside in the order from the starting point's projection point to the ending point's projection point refers to the path formed by sequentially connecting the starting point's projection point, the polyline vertices in the roadside between the starting point's projection point and the ending point's projection point along the roadside's direction. For example, if the sequence of polyline vertices of the roadside is... The starting projection point is The final projection point is The path obtained by trimming the roadside is: .
[0049] In one possible implementation, such as Figure 6If the starting point and the ending point are not on the same road, as shown in the diagram, path planning is performed in the road network topology based on the source and sink points determined in step S1300, in which both the initial cost and the tail cost are included in the calculation.
[0050] Specifically, firstly, each source point determined in step S1300, along with its corresponding initial cost, is placed into a pre-defined priority queue. The initial value of the current distance of a source point in the priority queue is the initial cost corresponding to that source point. Then, a graph search algorithm is executed in the road network topology. The graph search algorithm can be either Dijkstra's algorithm or A* algorithm. The graph search algorithm takes the priority queue as input, and in each iteration, it pops the node with the smallest current distance from the priority queue for processing until the priority queue is empty.
[0051] The output of the graph search algorithm is the shortest path from each source node to each sink node in the road network topology, along with the current distance obtained through that shortest path. After obtaining the shortest path, the current distance obtained through that shortest path needs to be updated using the tail segment cost of the sink node corresponding to that shortest path. One possible implementation is to sum the current distance of the shortest path from the source node to a sink node with the tail segment cost of that sink node to update the current distance of the shortest path.
[0052] For example, the source vertex is obtained through a graph search algorithm. to the remittance point The current distance of the shortest path is 10, and the sink is... The tail segment cost is 5, and the source point to the remittance point The current distance of the shortest path is 7, and the sink is... If the tail segment cost is 9, then the current distance of each shortest path is updated, thus finally obtaining the source point. to the remittance point The current distance of the shortest path is 15, and the source point... to the remittance point The current distance of the shortest path is 16. Finally, the shortest path with the smallest current distance, along with the node positions of the starting point and the ending point in the road network topology from the planning request, are selected to generate the final planned path.
[0053] It should be noted that the final planned path is a path consisting of the connection segment from the starting point of the planning request to the node location of the starting point in the road network topology, the connection segment from the node location of the starting point in the road network topology to the node location of the ending point in the road network topology in the current shortest path with the smallest distance, and the connection segment from the ending point to the node location of the ending point in the road network topology. For example, the starting point of the planning request is... The destination is The starting point is located at the node in the road network topology. The node location of the endpoint in the road network topology is The shortest path with the smallest current distance is ,in Located on the side of the road superior, Located on the side of the road The final planned path is as follows: .
[0054] <Device Embodiment> Figure 7 A schematic block diagram of a path planning apparatus according to an embodiment of the present disclosure is shown. Figure 7 As shown, the device 100 includes: Road network topology construction module 110: Determines the area where the current path to be planned is located, obtains the corresponding road network file based on the area, parses the road network file to generate the road network topology corresponding to the area; Access location determination module 120: Determines the start and end points of the current planned path from the received planning request, and determines the node position of the start point in the road network topology and the node position of the end point in the road network topology; Access cost calculation module 130: determines the initial cost based on the starting point and the node position of the starting point in the road network topology, and determines the tail segment cost based on the ending point and the node position of the ending point in the road network topology; Global Path Planning Module 140: Performs path planning based on the location of the starting point in the road network topology, the location of the ending point in the road network topology, the initial cost, and the tail cost, and uses the result of the path planning as the final planned path.
[0055] <Equipment Example> Figure 8 A schematic block diagram of a path planning device according to an embodiment of the present disclosure is shown. Figure 8 As shown, the path planning device 200 includes a processor 210 and a memory 220 for storing executable instructions of the processor 210. The processor 210 is configured to implement any of the path planning methods described above when executing the executable instructions.
[0056] It should be noted here that the number of processors 210 can be one or more. Furthermore, the path planning device 200 in this embodiment may also include an input device 230 and an output device 240. The processors 210, memory 220, input device 230, and output device 240 can be connected via a bus or other means, without specific limitations here.
[0057] The memory 220, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the path planning method in this embodiment of the present disclosure. The processor 210 executes various functional applications and data processing of the path planning device 200 by running the software program or module stored in the memory 220.
[0058] Input device 230 can be used to receive input digital numbers or signals. These signals may include key signals related to user settings and function control of the device / terminal / server. Output device 240 may include a display device such as a screen.
[0059] <Storage Medium Examples> According to a fourth aspect of this disclosure, a non-volatile computer-readable storage medium is also provided, on which computer program instructions are stored, which, when executed by processor 210, implement any of the path planning methods described above.
[0060] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technical improvements to the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.
Claims
1. A robot global path planning method based on road network topology, characterized in that, include: Determine the region where the current path to be planned is located, obtain the corresponding road network file based on the region, and parse the road network file to generate the road network topology corresponding to the region; The starting point and ending point of the current path to be planned are determined from the received planning request, and the node position of the starting point and the node position of the ending point in the road network topology are determined. The initial cost is determined based on the starting point and the node position of the starting point in the road network topology, and the tail segment cost is determined based on the ending point and the node position of the ending point in the road network topology; Based on the location of the starting point in the road network topology, the location of the ending point in the road network topology, the initial cost, and the tail cost, path planning is performed, and the result of the path planning is used as the final planned path.
2. The method according to claim 1, characterized in that, The step of parsing the road network file to generate the road network topology corresponding to the region includes: Nodes and road edges are parsed from the road network file; A basic adjacency table is constructed based on the adjacency relationships between each node and each road edge, and the road network topology is constructed based on the information recorded in the basic adjacency table.
3. The method according to claim 2, characterized in that, Determining the node position of the starting point and the node position of the ending point in the road network topology includes: Determine the starting point access location conditions and the ending point location constraints based on the starting point and the ending point, respectively. Each node and each roadside that satisfies the starting point access location condition and the ending point location constraint condition is respectively identified as a candidate node and candidate roadside for the starting point and a candidate node and candidate roadside for the ending point. The node position of the starting point in the road network topology is determined based on the candidate nodes and candidate roadsides of the starting point, and the node position of the ending point in the road network topology is determined based on the candidate nodes and candidate roadsides of the ending point.
4. The method according to claim 3, characterized in that, Determining the node position of the starting point in the road network topology based on candidate nodes and candidate road edges of the starting point, and determining the node position of the ending point in the road network topology based on candidate nodes and candidate road edges of the ending point, includes: The node position of the starting point in the road network topology is determined based on the distance between the starting point and each candidate node of the starting point, and the distance between the starting point and each candidate roadside of the starting point. The node position of the endpoint in the road network topology is determined based on the distance between the endpoint and each candidate node of the endpoint, and the distance between the endpoint and each candidate roadside of the endpoint.
5. The method according to claim 2, characterized in that, The process of determining the initial cost based on the starting point and its node position in the road network topology, and determining the tail segment cost based on the ending point and its node position in the road network topology, includes: The distance between the starting point and the node position of the starting point in the road network topology is calculated as the initial cost; The distance between the endpoint and the node location of the endpoint in the road network topology is calculated as the tail segment cost.
6. The method according to claim 5, characterized in that, The path planning based on the position of the starting point in the road network topology, the position of the ending point in the road network topology, and the initial cost and the tail cost includes: If the node position of the starting point and the node position of the ending point in the road network topology are located on the same road edge, then the road edge is directly used as the result of the path planning. If the node positions of the starting point and the ending point in the road network topology are not on the same road, then the source point and sink point are determined, and path planning that includes both the initial cost and the tail cost is executed, wherein the source point refers to the starting point for executing the path planning, and the sink point refers to the ending point for executing the path planning.
7. A robot global path planning device based on road network topology, characterized in that, include: Road network topology construction module: Determines the region where the current path to be planned is located, obtains the corresponding road network file based on the region, parses the road network file to generate the road network topology corresponding to the region; Access location determination module: Determines the start and end points of the current path to be planned from the received planning request, and determines the node position of the start point in the road network topology and the node position of the end point in the road network topology; Access cost calculation module: determines the initial cost based on the starting point and the node position of the starting point in the road network topology, and determines the tail segment cost based on the ending point and the node position of the ending point in the road network topology; Global Path Planning Module: Based on the position of the starting point in the road network topology, the position of the ending point in the road network topology, the initial cost, and the tail cost, perform path planning and use the result of the path planning as the final planned path.
8. The apparatus according to claim 7, characterized in that, The road network topology construction module parses the nodes and road edges from the road network file, constructs a basic adjacency table based on the adjacency relationship between each node and each road edge, and builds the road network topology based on the information recorded in the basic adjacency table.
9. A robot global path planning device based on road network topology, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 6 when executing the executable instructions.
10. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.
Citation Information
Patent Citations
A transport vehicle scheduling method for a smart port
CN120952496B
