Four-way shuttle robot intelligent path planning method

Abstract

The application belongs to the technical field of path planning, and particularly relates to an intelligent path planning method for a four-way shuttle robot, which comprises the following steps: an initial path is generated by using a JPS algorithm, and key anchor points are extracted; when the robot encounters a dynamic obstacle during driving, the system responds through an innovative hierarchical decision mechanism: firstly, the stability of the anchor point behind the conflict point is evaluated, if the stability is high, a minimum repair window is used, if the stability is uncertain or low, a topological influence depth verification is started to dynamically define an optimal local repair window, finally, local path re-planning is carried out in the window, and the generated new path segment is seamlessly spliced with the original path. Through the adaptive local repair strategy, efficient and low-cost path updating in a dynamic environment is realized, and the stability and efficiency of the robot operation are ensured.

Description

Technical Field

[0001] This invention relates to the field of path planning technology. More specifically, this invention relates to an intelligent path planning method for a four-way shuttle robot. Background Technology

[0002] Four-way shuttle robots are the core execution units of modern dense warehousing systems. Their operating efficiency is directly related to the throughput and response speed of the entire logistics system. Path planning, as the brain of the robot, has the core task of quickly and accurately calculating an optimal or suboptimal path from the starting position to the target position in a complex gridded warehouse environment.

[0003] In real-world warehousing operations, the environment is not static. In addition to static layouts such as shelves and columns, there are often dynamic obstacles such as temporarily parked equipment, malfunctioning other robots, or temporarily stacked goods. These unexpected situations can render the pre-planned globally optimal path infeasible. In such cases, the robot must be able to intelligently adjust its route to avoid obstacles. This places extremely high demands on the dynamic adaptability of path planning algorithms.

[0004] Currently, the common technical approach to solving such problems is to use graph search-based path planning algorithms, such as the classic A-Star algorithm or its efficient improved version on grid maps, namely the JPS (Jump Point Search) algorithm. Given a static map, such algorithms can efficiently calculate a globally optimal path.

[0005] However, when the path encounters dynamic obstacles during execution, the most direct response of existing technologies is to trigger a complete path replanning, that is, to recalculate a completely new path to the destination from the robot's current position. This complete replanning strategy has significant drawbacks: First, the global search computational overhead is huge, which will cause the robot to pause and wait for a long time, seriously affecting the work efficiency; second, the newly generated path may be structurally very different from the original path. Even if it is only to avoid a small local obstacle, it may cause the path to shake violently, destroying the stability and predictability of the path, and adversely affecting the collaborative scheduling and operational stability of the entire multi-robot system. Summary of the Invention

[0006] To address the technical problems in existing technologies where robots encounter dynamic obstacles, either global path replanning leads to high computational load and long robot pause times, or the use of fixed repair areas results in low repair success rates or insufficient efficiency, this invention provides an intelligent path planning method for a four-way shuttle robot. The method includes: acquiring a rasterized map containing a static environment layout; calculating an initial optimal path from the starting position to the target position based on the map using the JPS algorithm, and extracting multiple jump points, the starting position, and the target position from the initial optimal path into an ordered sequence of path anchor points; during the robot's movement along the initial optimal path, in response to the detection of a dynamic obstacle, determining whether the obstacle conflicts with the initial optimal path: if a conflict exists... If a conflict occurs, a stability assessment is performed on the first anchor point after the conflict point to obtain its stability assessment index. The stability assessment index is directly proportional to the distance from the first anchor point to the obstacle and inversely proportional to the distance from the first anchor point to the robot's current position, and increases as its position in the remaining path progresses. Based on the comparison result of the stability assessment index and a preset threshold, a hierarchical decision is made to define a repair window containing the conflict point. The boundary of the repair window is determined by two anchor points in the path anchor point sequence. A local path search is performed within the repair window to generate an optimal local repair path that bypasses the obstacle. The optimal local repair path is then spliced ​​with the unaffected path segments in the initial optimal path to form a complete new path.

[0007] This invention pre-extracts path anchor points and performs multi-dimensional stability assessment on the first anchor point after a conflict, dynamically adjusting the repair window to narrow the scope of path replanning from the entire map to the local area affected by obstacles. This avoids a complete recalculation of the entire path, and in particular avoids repeated searches for a large number of still valid path segments, thereby significantly reducing the computational complexity and time consumption of the algorithm and enabling it to respond to dynamic changes in the environment at a faster speed.

[0008] Furthermore, the anchor point stability evaluation index designed in this invention can dynamically determine the impact range of obstacles and define the repair area, so that path repair is only performed locally. The newly generated path maintains maximum consistency with the original path, effectively avoiding the problem of severe path jitter caused by global recalculation. This makes the robot's running trajectory smoother and more predictable. In high-density multi-robot collaborative operation scenarios, it reduces the chain interference caused by drastic changes in individual paths to other surrounding robots, improving the operational stability and overall operational efficiency of the entire warehousing system. At the same time, by combining the inherent speed advantage of the JPS algorithm with the idea of ​​incremental repair, the path planning system can respond quickly and make high-quality, low-impact decisions when faced with sudden failures or obstacles, significantly enhancing the robustness and fault tolerance of the entire automated warehousing system.

[0009] Preferably, determining that the obstacle conflicts with the initial optimal path includes: assuming the robot's current position coordinates are... The paths that have not been traversed are those from the initial optimal path. Sub-path to target location; check obstacle coordinates Is it located on the untraveled sub-path? If If the conflict exists on the sub-path, a path conflict is determined, and the distance on the path is recorded. The system identifies the nearest conflict point and locates the anchor point interval containing the conflict point within the path anchor point sequence. .

[0010] Preferably, a stability assessment is performed on the first anchor point after the conflict point to obtain its stability assessment index, including: As the first anchor point after the conflict point, its stability assessment index The formula for calculation is: In the formula, The Manhattan distance between the two points; The coordinates of the dynamic obstacle; These are the robot's current position coordinates; It is the last effective anchor point before the conflict occurs; From the anchor point From the initial optimal path to the anchor point Path length; From the anchor point From the initial optimal path to the destination The total remaining path length.

[0011] This invention comprehensively considers three dimensions—the distance between the anchor point and the obstacle, the distance to the robot's current position, and the relative position in the remaining path—to calculate the anchor point stability evaluation index. This makes the judgment of anchor point stability more accurate and comprehensive, providing a data foundation for the reliability of subsequent hierarchical decision-making and avoiding repair failures or inefficiencies caused by misjudgments based on a single dimension.

[0012] Preferably, the preset threshold includes a high stability threshold. and low stability threshold ,and .

[0013] Preferably, a hierarchical decision is made based on the comparison result between the stability evaluation index and the preset threshold, defining a repair window that includes the conflict point, including: if the anchor point Stability assessment indicators Then repair the starting anchor point of the window. End point anchor ;if The system initiates topology impact depth verification to determine the repair window.

[0014] Based on a set dual threshold, this invention constructs a hierarchical decision-making process: for anchor points with extremely high stability, the minimum repair window is adopted to pursue the highest efficiency; for situations with uncertain stability, a more in-depth verification is initiated. This hierarchical processing mechanism avoids a one-size-fits-all approach to all conflicts, significantly improving the overall efficiency and intelligence level of the system in dealing with dynamic obstacles of varying complexity.

[0015] Preferably, the step of initiating topology impact depth verification to determine the repair window includes: using Starting from, with For the first exploration endpoint, perform a bounded breadth-first search: if a path to the exploration endpoint can be found in the bounded breadth-first search... The path then repairs the starting anchor point of the window. End point anchor If, in a bounded breadth-first search, a path cannot be found within the specified range... The path needs to be expanded to include a wider window: maintain the starting anchor point of the repair window. The exploration endpoint will remain unchanged, and will be moved forward to the next anchor point. The stability assessment is repeated, and a tiered decision is made based on the comparison between the stability assessment index and the preset threshold, until the first one that can be [performed / resolved] is found. Connected anchor points Then repair the end anchor point of the window. .

[0016] The present invention introduces a topology influence depth verification mechanism that is initiated when the stability of anchor points is uncertain. When simple geometric evaluation is insufficient to make a judgment, the method uses a bounded breadth-first search to detect the actual path connectivity and can dynamically expand the repair window when necessary. This solves the problem that existing technologies may fall into the topology trap of seemingly distant but actually necessary paths being blocked if they rely solely on distance judgment. This greatly improves the success rate and robustness of path repair in complex congestion scenarios.

[0017] Preferably, when performing a bounded breadth-first search, the search range is limited to a range of [missing information]. and Within a circular region with a radius equal to the Manhattan distance between them.

[0018] This invention limits the breadth-first search range by confining the search area to a circular region with a radius equal to the Manhattan distance, effectively controlling the computational complexity of the topology verification process. This ensures verification validity while avoiding a surge in computational load caused by an infinitely expanding search range, achieving a balance between higher success rate and computational efficiency.

[0019] Preferably, performing a local path search within the repair window includes: using As the starting point for local search, with As the endpoint of the local search, after updating the dynamic obstacles... On the map copy, a route from... is calculated using the JPS algorithm. arrive And it can bypass obstacles Optimal local repair path .

[0020] Preferably, the new path consists of three parts: the original path starting from the origin... To the starting point of repair sub-path segments The newly generated optimal local repair path And the original path from the repair endpoint To the finish line sub-path segments .

[0021] Preferably, the jump points include corner points and / or obstacle proximity points in the initial optimal path.

[0022] The beneficial effects of this invention are as follows:

[0023] This invention pre-extracts path anchor points and performs multi-dimensional stability assessment on the first anchor point after a conflict, dynamically adjusting the repair window to narrow the scope of path replanning from the entire map to the local area affected by obstacles. This avoids a complete recalculation of the entire path, and in particular avoids repeated searches for a large number of still valid path segments, thereby significantly reducing the computational complexity and time consumption of the algorithm and enabling it to respond to dynamic changes in the environment at a faster speed.

[0024] Furthermore, the anchor point stability evaluation index designed in this invention can dynamically determine the impact range of obstacles and define the repair area, so that path repair is only performed locally. The newly generated path maintains maximum consistency with the original path, effectively avoiding the problem of severe path jitter caused by global recalculation. This makes the robot's running trajectory smoother and more predictable. In high-density multi-robot collaborative operation scenarios, it reduces the chain interference caused by drastic changes in individual paths to other surrounding robots, improving the operational stability and overall operational efficiency of the entire warehousing system. At the same time, by combining the inherent speed advantage of the JPS algorithm with the idea of ​​incremental repair, the path planning system can respond quickly and make high-quality, low-impact decisions when faced with sudden failures or obstacles, significantly enhancing the robustness and fault tolerance of the entire automated warehousing system. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating an intelligent path planning method for a four-way shuttle robot according to the present invention;

[0026] Figure 2 This is a flowchart illustrating step S3;

[0027] Figure 3 It is a schematic map showing the warehouse information;

[0028] Figure 4 It is a schematic diagram showing the initial optimal path;

[0029] Figure 5 This is a schematic diagram illustrating the planning of a new path for local path repair using the method of the present invention;

[0030] Figure 6 It is a schematic diagram showing the planning of a completely new path recalculated using existing technology. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0033] This invention discloses an intelligent path planning method for a four-way shuttle robot, referring to... Figure 1 This includes steps S1 to S4:

[0034] S1: Generate the initial path and extract the path anchors.

[0035] It should be noted that in order to perform local and incremental repairs when path conflicts occur, rather than recalculating globally, the initially generated complete path first needs to be structured to extract the key nodes that can represent the path skeleton, i.e., path anchor points.

[0036] Specifically, the system acquires the gridded map information of the warehouse and the starting position coordinates of the four-way shuttle robot. and the coordinates of the mission target The map information includes static environmental layouts such as shelves and aisles.

[0037] Furthermore, the system executes the standard JPS algorithm once to calculate a route from... arrive Initial optimal path The path consists of a series of continuous grid coordinate points.

[0038] Furthermore, to facilitate subsequent local repairs, starting from the initial optimal path... The algorithm identifies all jump points, which are key decision points identified by the JPS algorithm during the search process, including but not limited to path corners and obstacle proximity points.

[0039] Finally, list all jump points and the starting points of the paths. and the end point Combined, they form an ordered sequence of path anchor points. ,in and .

[0040] S2: Perform dynamic obstacle conflict detection.

[0041] It should be noted that during the robot's movement, environmental changes must be monitored in real time. Once a dynamic obstacle appears, it should be immediately determined whether it poses a threat to the current planned path. This is a prerequisite for activating the path repair mechanism.

[0042] Specifically, as the robot follows the initial optimal path During the journey, the system monitors the warehouse environment in real time: when it receives information about dynamic obstacles, such as a malfunctioning robot reporting its location coordinates... When an obstacle appears, the system immediately determines whether it conflicts with the currently planned path, including:

[0043] Let the robot's current position coordinates be... The path that has not been traversed is the initial optimal path. Zhong Cong arrive Sub-paths; check obstacle coordinates Is it located on the untraveled sub-path? If If the conflict exists on the sub-path, a path conflict is determined, and the distance on the path is recorded. Recent points of conflict Meanwhile, the system in the path anchor point sequence Locating the point of conflict The anchor point range .

[0044] S3: In response to the detection of a dynamic obstacle that conflicts with the initial optimal path, a stability assessment is performed on the first anchor point after the conflict point. Based on the comparison between the stability assessment index and the preset threshold, a graded decision is made to define the repair window.

[0045] It should be noted that, in order to minimize computational overhead while ensuring a high success rate of repairs, a fixed-size repair window cannot be used. Instead, the window range should be adaptively determined based on the actual degree of the obstacle's impact. Therefore, this step achieves this goal through a hierarchical process that combines low-overhead geometric evaluation and high-precision topology verification.

[0046] Refer to the flowchart for step S3 Figure 2 This includes steps S301 to S303, specifically:

[0047] S301: Perform a stability assessment on the first anchor point after the point of conflict.

[0048] It is important to note that a key issue in determining the boundaries of the repair window is identifying which anchor point after the conflict point is stable and unaffected by obstacles, thus serving as a reliable repair endpoint. The stability of an anchor point depends not only on its physical distance from the obstacle but also on its importance and location in future path planning; therefore, constructing a comprehensive evaluation index is a prerequisite for scientifically defining the window.

[0049] Specifically, the system sets the first anchor point after the conflict point. Conduct a stability assessment, and its stability assessment indicators Calculate using the following formula:

[0050]

[0051] In the formula, anchor point Stability assessment indicators; The Manhattan distance between the two points reflects the actual travel distance on the raster map; The coordinates of the dynamic obstacle; These are the robot's current position coordinates; It is the last effective anchor point before the conflict occurs; From the anchor point From the initial optimal path to the anchor point Path length; From the anchor point From the initial optimal path to the destination The total remaining path length.

[0052] The first term of the calculation formula The obstacle was measured relative to the robot's current position relative to the anchor point. The relative influence; the second term of the calculation formula It is a weighting factor that gives higher stability weights to later anchor points; therefore, the calculation formula combines two dimensions: distance and path position. Distance to obstacles The farther away, or the further away from the current robot position The closer they are, the greater the ratio of the first term; at the same time, when the anchor point... The further along the remaining path the value is, the larger the logarithm of the second term becomes; the combined effect of both makes... The larger the value, the higher the anchor point. The more stable it is, the less likely it is to be affected by this dynamic obstacle.

[0053] It should be noted that this indicator can quickly assess the degree of impact on anchor points, providing data for subsequent hierarchical decisions and avoiding the one-sidedness of judging solely based on a single distance.

[0054] S302: Make tiered decisions based on stability assessment results.

[0055] It should be noted that, in order to balance decision-making efficiency and accuracy, different processing strategies are adopted for anchor points with different levels of stability. For extremely stable cases, the minimum window is used directly to pursue the highest efficiency; for uncertain cases, a more precise but more expensive verification process is adopted.

[0056] Specifically, a high stability threshold is preset. and a low stability threshold ,and High stability threshold and low stability threshold Its function is to divide the decision-making area; therefore, in this embodiment, it will... Set to 0.8, Setting it to 0.3 means that when the relative distance between an obstacle and an anchor point is more than 4 to 5 times the relative distance between the robot and the anchor point, the system tends to classify it as a high-stability region; while when the distance between the two is close or smaller, it tends to classify it as an unstable region, thereby initiating more in-depth topology verification and achieving a balance between efficiency and decision reliability. In other embodiments, implementers can adjust these two thresholds according to the actual situation such as the complexity of the warehouse layout and the density of robots.

[0057] Furthermore, based on the anchor point Stability assessment indicators The relationship with the threshold is determined by a tiered decision-making process in three cases:

[0058] Scenario 1: If the anchor point Stability assessment indicators The system determines the anchor point. Minimal impact from obstacles ensures a very safe geometric position, eliminating the need for complex topology analysis. The system directly determines the repair window as the minimum range, thus establishing the starting anchor point of the repair window. End point anchor This method constructs a fast path that can handle the vast majority of simple local obstacles, making it the most efficient.

[0059] Scenario 2: If This indicates that the stability assessment results of the anchor points are in a gray area, and the anchor points The stability of the system is uncertain; at this point, simple geometric distance is not enough to make a reliable judgment. There may be topological traps that seem far away but are actually inevitable. Therefore, the system must initiate topological influence depth verification to determine the repair window.

[0060] Scenario 3: If This indicates that, geometrically, the anchor point It is already very unstable; at this point, the system also needs to initiate topology impact depth verification to determine the repair window in order to find the truly stable boundary.

[0061] S303: In response to uncertain or low stability, initiate topology impact depth verification to define the repair window.

[0062] It should be noted that when geometric assessment cannot provide a definitive conclusion, the actual connectivity of the map, i.e., the topological structure, must be used to determine the true extent of the obstacle's impact. This verification can avoid being misled by the appearance of geometric distance, such as a seemingly distant but necessary narrow passage being blocked.

[0063] Specifically, when entering situation two or three, with Starting from, with For the first exploration endpoint, a bounded breadth-first search is performed to attempt to find a path to the current exploration endpoint anchor point. The search area is limited to a relatively small region, i.e., within a certain range. and Within a circular region with a radius equal to the Manhattan distance between them, the computational cost is strictly limited.

[0064] Alternatively, in a bounded breadth-first search, if a path leading to the exploration endpoint can be found quickly... The path indicates that the obstacle only causes a local detour and does not disrupt the critical topology; at this point, the system sets the repair window to the minimum range, and the starting anchor point of the repair window... End point anchor .

[0065] Optionally, if a bounded breadth-first search fails to find a path within the specified range... The path indicates that the obstacle blocked it. and The necessary passage between them creates a local topological bottleneck; at this point, the system determines that the obstacle has a significant impact, requiring an expansion of the window range: maintaining the starting anchor point of the repair window. for The exploration endpoint will remain unchanged, and will be moved forward to the next anchor point. The stability assessment is repeated, and a tiered decision is made based on the comparison between the stability assessment index and the preset threshold, until the first one that can be [performed / resolved] is found. Connected anchor points The anchor point It is then ultimately determined as the endpoint anchor point of the repair window. This ensures that the endpoint of the repair window is located in an open area outside the actual topological influence range of the obstacle, thereby guaranteeing the success rate and path quality of the local JPS search.

[0066] It should be noted that, in this way, the size of the repair window no longer depends on a potentially inaccurate distance calculation formula, but is dynamically determined by the degree to which obstacles disrupt the actual connectivity of the map, ensuring that the boundaries of the repair window are reliable in complex situations.

[0067] S4: Perform a local path search within the repair window to generate the optimal local repair path, and then merge it with the unaffected path segments in the initial optimal path to form a complete new path.

[0068] It should be noted that once a reliable repair boundary is determined, efficient path searching can be performed within the restricted area, and the conflicting parts of the original path can be replaced with the newly generated local path, thereby completing the entire repair process.

[0069] First, within the defined repair window, the system performs a local JPS path search: As the starting point for local search, with As the endpoint of the local search, after updating the dynamic obstacles... On the map copy, calculate a route from arrive And it can bypass obstacles Optimal local repair path .

[0070] Then, the newly generated optimal local repair path Seamlessly splice it with the stable path segments preserved in the original path to form a complete and restored new path. New Path It consists of three parts: the original path from the starting point To the starting point of repair sub-path segments The newly generated optimal local repair path And the original path from the repair endpoint To the finish line sub-path segments .

[0071] Finally, the system will use this new path The task was assigned to the four-way shuttle robot for execution.

[0072] It should be noted that since the new path only changes in a local area and maintains a high degree of consistency with the main structure of the original path, it avoids severe path shaking and ensures the stability and predictability of the robot's operation.

[0073] For example, the map information of the warehouse is as follows: Figure 3 As shown, this includes the starting position and the mission objective position; for Figure 3 The system uses the warehouse map information to execute a standard JPS algorithm to calculate an initial optimal path from the starting position to the task target position. The initial optimal path planning diagram is shown below. Figure 4 As shown; during the robot's journey along the initial optimal path, the system monitors the warehouse environment in real time. When the robot reaches... Figure 4 Upon reaching the current position, a report and coordinates of a malfunctioning robot are received, and it is determined that the obstacle conflicts with the currently planned path. Further, the method of this invention is used to repair the local path, and the planned diagram of the repaired new path is shown below. Figure 5 As shown; while existing technology recalculates a completely new path to the task target location from the robot's current position, as shown in the diagram of the new path. Figure 6 As shown; compared to Figure 6 A completely new path in the process, Figure 5 The repaired path only changes in local areas, maintaining a high degree of consistency with the main structure of the original path, thus avoiding severe path jitter and ensuring the smoothness and predictability of robot operation.

Claims

1. A method for intelligent path planning of a four-way shuttle robot, characterized in that, include: Obtain a rasterized map containing the static environment layout; Based on the map, the JPS algorithm is used to calculate the initial optimal path from the starting position to the target position. Multiple jump points, starting positions, and target positions in the initial optimal path are extracted into an ordered sequence of path anchor points. During the robot's journey along the initial optimal path, in response to the detection of dynamic obstacles, it is determined that the obstacle conflicts with the initial optimal path. If a conflict exists, a stability assessment is performed on the first anchor point after the conflict point to obtain its stability assessment index, including: As the first anchor point after the conflict point, its stability assessment index The formula is: In the formula, The Manhattan distance between the two points; The coordinates of the dynamic obstacle; These are the robot's current position coordinates; It is the last effective anchor point before the conflict occurs; From the anchor point From the initial optimal path to the anchor point Path length; From the anchor point From the initial optimal path to the destination The total remaining path length, the stability evaluation index is directly proportional to the distance from the first anchor point to the obstacle, inversely proportional to the distance from the first anchor point to the robot's current position, and increases as its position in the remaining path becomes more advanced; A hierarchical decision is made based on the comparison results between the stability assessment index and the preset threshold, and a repair window containing conflict points is defined. The boundary of the repair window is determined by two anchor points in the path anchor point sequence. Perform a local path search within the repair window to generate an optimal local repair path that bypasses the obstacle; then merge the optimal local repair path with the unaffected path segments in the initial optimal path to form a complete new path.

2. The intelligent path planning method for a four-way shuttle robot according to claim 1, characterized in that, Determining if an obstacle conflicts with the initial optimal path includes: Let the robot's current position coordinates be... The paths that have not been traversed are those from the initial optimal path. Subpath to the target location; Check obstacle coordinates Is it located on the untraveled sub-path? If If the conflict exists on the sub-path, a path conflict is determined, and the distance on the path is recorded. The system identifies the nearest conflict point and locates the anchor point interval containing the conflict point within the path anchor point sequence. .

3. The intelligent path planning method for a four-way shuttle robot according to claim 1, characterized in that, Preset thresholds include high stability thresholds. and low stability threshold ,and .

4. The intelligent path planning method for a four-way shuttle robot according to claim 3, characterized in that, Based on the comparison results between stability assessment indicators and preset thresholds, a hierarchical decision is made, defining a repair window that includes conflict points, including: If anchor point Stability assessment indicators Then repair the starting anchor point of the window. End point anchor ; if The system initiates topology impact depth verification to determine the repair window.

5. The intelligent path planning method for a four-way shuttle robot according to claim 4, characterized in that, Initiate topology impact depth verification to determine the repair window, including: by Starting from, with For the first exploration endpoint, a bounded breadth-first search is performed: If, in a bounded breadth-first search, a path leading to the exploration endpoint can be found... The path then repairs the starting anchor point of the window. End point anchor ; If, in a bounded breadth-first search, a path cannot be found within the specified range... The path needs to be expanded to include a wider window: maintain the starting anchor point of the repair window. The exploration endpoint will remain unchanged, and will be moved forward to the next anchor point. The stability assessment is repeated, and a tiered decision is made based on the comparison between the stability assessment index and the preset threshold, until the first one that can be [performed / resolved] is found. Connected anchor points Then repair the end anchor point of the window. .

6. The intelligent path planning method for a four-way shuttle robot according to claim 5, characterized in that, When performing a bounded breadth-first search, the search scope is limited to a range of elements. and Within a circular region with a radius equal to the Manhattan distance between them.

7. The intelligent path planning method for a four-way shuttle robot according to claim 1, characterized in that, Perform a local path search within the repair window, including: by As the starting point for local search, with As the endpoint of the local search, after updating the dynamic obstacles... On the map copy, a route from... is calculated using the JPS algorithm. arrive And it can bypass obstacles Optimal local repair path .

8. The intelligent path planning method for a four-way shuttle robot according to claim 1, characterized in that, The new path consists of three parts: the original path starting from the origin To the starting point of repair sub-path segments The newly generated optimal local repair path And the original path from the repair endpoint To the finish line sub-path segments .

9. The intelligent path planning method for a four-way shuttle robot according to claim 1, characterized in that, Jump points include corner points and / or points near obstacles in the initial optimal path.

Images