Robot getting-off task execution logic control method

Abstract

The invention provides a robot getting-off task execution logic control method which comprises the steps that when a task is executed or a new task is received and needs to be returned to the other side of an obstacle, a robot is controlled to move to a recorded getting-off position; judging whether the trolley is still located at the getting-off position; if yes, the robot is controlled to execute the trolley loading action, and the robot continues to move after being converted into the carrying state; if not, further judging whether the current task is an emergency task or not; if the current task is a non-emergency task, the robot is controlled to find a trolley in the view field, if the trolley is found, the robot continues to move after executing the action of getting on the trolley, and if the trolley is not found, the robot is controlled to walk to execute follow-up movement; and if the current task is an emergency task, the robot is controlled to give up searching for the trolley and directly walk to execute subsequent movement. According to the invention, the overall task execution efficiency and reliability of the composite robot system in a complex and dynamic environment are improved.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, and in particular to a logic control method for a robot to perform tasks after disembarking. Background Technology

[0002] Wheeled mobile robots and bipedal humanoid robots are two important robot forms, each with its own distinct advantages and limitations. Wheeled robots (or those mounted on platform vehicles) typically offer advantages such as smooth movement, high speed, low energy consumption, and relatively simple control, exhibiting high mobility in flat or structured environments. In contrast, bipedal humanoid robots possess stronger terrain adaptability and anthropomorphic interaction capabilities, enabling them to traverse obstacles such as steps and ditches, making them irreplaceable in unstructured environments or scenarios requiring seamless integration with human environments.

[0003] To balance mobility and obstacle-crossing capabilities, some existing technologies attempt to combine the two. For example, equipping bipedal robots with wheeled platforms for mobility aims to leverage the platform's efficiency during long-distance, flat movements, allowing the robot to dismount and perform complex actions when needed. However, current solutions often focus on mechanical design or simple load-unload patterns, lacking a set of intelligent collaborative control logic deeply coupled with complex task scenarios.

[0004] Specifically, existing solutions have the following shortcomings: First, during the movement from the starting point to the target point, when encountering obstacles, there is a lack of a dynamic evaluation mechanism to determine whether it is better for the robot-vehicle assembly to directly bypass the obstacle or for the robot to dismount and walk over it. A fixed strategy is often adopted, leading to low efficiency or even impassability. Second, existing technologies do not adequately consider how to efficiently and safely return and rejoin the vehicle after the robot dismounts to perform its task. In particular, when the vehicle is not at its original dismount position for any reason (such as being moved by someone else), the robot lacks an effective response strategy and may be trapped in a predicament of blindly searching or task interruption. Furthermore, existing solutions fail to flexibly adjust the robot's strategy for finding the vehicle according to the urgency of the task, failing to achieve a balance between task timeliness and system energy consumption. Summary of the Invention

[0005] The purpose of this application is to provide a logic control method for robots to perform tasks after disembarking, thereby improving the overall task execution efficiency and reliability of composite robot systems in complex and dynamic environments.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] A robot disembarking and performing a task logic control method, the control method comprising:

[0008] Step 1: In response to the task trigger, control the robot in the carrying state and its carrying vehicle to move together towards the target point; during the movement, if an obstacle is detected in front, determine whether the obstacle can be bypassed by the vehicle; if the determination is yes, control the vehicle to bypass the obstacle and continue moving towards the target point; if the determination is no, control the vehicle to stop at a preset convenient location, record the convenient location as the disembarkation location, control the robot to disembark and walk across the obstacle to the target point to perform the task;

[0009] Step 2: When the task is completed or a new task is received that requires returning to the other side of the obstacle, control the robot to move to the recorded disembarkation position; determine whether the car is still in the disembarkation position; if the determination is yes, control the robot to perform the car mounting action, switch to the mounted state and continue moving; if the determination is no, further determine whether the current task is an emergency task.

[0010] Step 3: If the current task is not urgent, control the robot to search for the car within its field of vision. If it finds the car, it will perform the action of getting on the car and continue moving. If it does not find the car, control the robot to walk to perform the subsequent movement.

[0011] Step 4: If the current task is an emergency task, control the robot to abandon the search for the car and proceed on foot to carry out the subsequent movement.

[0012] The robot's task execution logic control system after disembarking includes:

[0013] The control module is used to respond to task triggering and control the robot and its mounted vehicle to move together toward the target point. During the movement, if an obstacle is detected in front, it determines whether the obstacle can be bypassed by the vehicle. If the determination is yes, it controls the vehicle to bypass the obstacle and continue moving toward the target point. If the determination is no, it controls the vehicle to stop at a preset convenient location and records the convenient location as the disembarkation location. It then controls the robot to disembark and walk across the obstacle to the target point to perform the task.

[0014] The judgment module is used to control the robot to move to the recorded disembarkation position when it has completed a task or received a new task that requires it to return to the other side of the obstacle; to determine whether the vehicle is still in the disembarkation position; if the determination is yes, the robot is controlled to perform the action of getting on the vehicle, and continues to move after entering the carrying state; if the determination is no, the robot is further determined to determine whether the current task is an emergency task.

[0015] The execution module is used to control the robot to search for a car within its field of vision if the current task is not urgent. If a car is found, the robot will mount the car and continue moving. If a car is not found, the robot will walk to continue moving. If the current task is urgent, the robot will abandon the search for a car and walk directly to continue moving.

[0016] Beneficial effects:

[0017] This invention dynamically determines whether the vehicle can circumvent obstacles during movement, deciding whether to employ a combined vehicle-mounted approach or a robot-dismounted obstacle-crossing strategy. When circumvention is feasible, efficient vehicle-mounted movement is prioritized; when circumvention is not feasible, the system promptly switches to robot obstacle-crossing mode. This dynamic switching mechanism enables the composite robot system to automatically select the optimal movement method under different terrain conditions, thereby maximizing the overall movement efficiency and task execution speed of the system while ensuring accessibility.

[0018] This invention features a robust return and re-boarding logic. After completing its task, the robot proactively returns to its recorded disembarkation location and first checks if the vehicle is still in its original position. This mechanism ensures that the return-boarding process can be completed quickly and accurately in most cases. More importantly, when the vehicle is found to be out of position, the system does not simply wait or report an error; instead, it introduces a tiered response strategy based on the urgency of the task, significantly enhancing the system's adaptability and robustness to dynamic disturbances (such as the vehicle being moved unexpectedly).

[0019] This invention creatively incorporates task attributes (urgent / non-urgent) into the control logic. For non-urgent tasks, the robot searches for the vehicle within a limited range, attempting to resume efficient movement, which helps conserve the robot's walking energy and extend its range. For urgent tasks, the search is abandoned decisively, prioritizing the timeliness of the task. This differentiated processing strategy allows the system to flexibly achieve the optimal balance between pursuing the most efficient movement and the fastest task response, based on the actual needs of the task. Attached Figure Description

[0020] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. Wherein:

[0021] Figure 1 A schematic diagram of the logic control method for the robot to dismount and perform tasks. Detailed Implementation

[0022] The present application will now be described in detail with reference to the accompanying drawings and embodiments. Various examples are provided by way of explanation and not by way of limitation. In fact, those skilled in the art will recognize that modifications and variations can be made to the present application without departing from the scope or spirit thereof. For example, a feature shown or described as part of one embodiment may be used in another embodiment to produce yet another embodiment. Therefore, it is desirable that the present application encompass such modifications and variations that fall within the scope of the appended claims and their equivalents.

[0023] like Figure 1 As shown, the robot dismounts to perform a task logic control method according to the present invention includes:

[0024] Step 1: In response to the task trigger, control the robot in the carrying state and its carrying vehicle to move together towards the target point; during the movement, if an obstacle is detected in front, determine whether the obstacle can be bypassed by the vehicle; if the determination is yes, control the vehicle to bypass the obstacle and continue moving towards the target point; if the determination is no, control the vehicle to stop at a preset convenient location, record the convenient location as the disembarkation location, control the robot to disembark and walk across the obstacle to the target point to perform the task;

[0025] Step 2: When the task is completed or a new task is received that requires returning to the other side of the obstacle, control the robot to move to the recorded disembarkation position; determine whether the car is still in the disembarkation position; if the determination is yes, control the robot to perform the car mounting action, switch to the mounted state and continue moving; if the determination is no, further determine whether the current task is an emergency task.

[0026] Step 3: If the current task is not urgent, control the robot to search for the car within its field of vision. If it finds the car, it will perform the action of getting on the car and continue moving. If it does not find the car, control the robot to walk to perform the subsequent movement.

[0027] Step 4: If the current task is an emergency task, control the robot to abandon the search for the car and proceed on foot to carry out the subsequent movement.

[0028] In this embodiment, a layered obstacle handling strategy that prioritizes detours and allows for necessary dismounts minimizes ineffective pauses during task execution. During the movement phase, the feasibility of detouring around obstacles is first assessed. If detouring is possible, the robot is directly controlled to detour and continue moving towards the target point, avoiding unnecessary robot dismount-walk-remount processes and saving task time. If the obstacle cannot be detouring, the robot stops at a pre-set convenient location and dismounts to walk over the obstacle, ensuring the task is not interrupted by obstacles. Simultaneously, upon task completion or when returning, the robot is first checked to ensure it is still in the dismounted position. If it is, a quick remounting action is performed to restore the mounted state, ensuring the efficiency of subsequent movement stages and effectively improving the overall continuity and efficiency of task execution.

[0029] This control method effectively combines the mobility of a vehicle with the obstacle-crossing capabilities of a robot, achieving excellent adaptability to various complex environments. While the vehicle possesses strong ground mobility, it faces limitations when encountering large obstacles or narrow passages. The robot, on the other hand, has the ability to overcome obstacles on foot, compensating for the vehicle's shortcomings in complex terrain. Through a collaborative mode of vehicle-mounted and robot-walking obstacle-crossing, the entire execution system can achieve rapid movement on flat surfaces using the vehicle, and when encountering obstacles that the vehicle cannot pass, the robot can dismount and walk to overcome them, effectively overcoming the limitations of a single movement method. Whether in open spaces, areas with scattered obstacles, or complex environments containing large obstacles that the vehicle cannot bypass, this method ensures smooth task progress, significantly expanding the scope of scenarios in which the robot can perform tasks.

[0030] Through refined decision-making logic, the rational utilization of resources for both the vehicle and the robot was achieved, reducing unnecessary energy consumption and equipment wear and tear. On the one hand, in the obstacle handling phase, the vehicle's detour was prioritized, avoiding the additional energy consumption caused by the robot frequently getting off, walking, and getting back on (typically, the robot's walking energy consumption is higher than the vehicle's movement energy consumption). On the other hand, in the return phase, for non-urgent tasks, the robot was controlled to search for the vehicle within its field of vision before performing the boarding action, avoiding the increased energy consumption caused by the robot blindly walking. Only in urgent task scenarios, to ensure task priority, the search for the vehicle was abandoned and the robot walked directly, achieving a balance between efficiency and resource conservation. At the same time, unnecessary mechanical docking between the robot and the vehicle was reduced, mechanical wear and tear during the docking process was decreased, and the equipment's service life was extended.

[0031] This method introduces a task priority judgment mechanism, which effectively ensures the smooth execution of urgent tasks in special scenarios and improves the system's emergency response level. When the robot returns to the disembarkation position but does not find the car, it further determines whether the current task is urgent: if it is urgent, the robot is directly controlled to abandon the search for the car and proceed on foot, ensuring that the urgent task can proceed quickly according to priority requirements and avoiding delays caused by searching for the car; if it is not urgent, the process of searching for the car is executed, taking into account resource utilization efficiency. This differentiated processing strategy based on task priority enables the system to dynamically adjust the execution logic according to the urgency of the task, meeting the execution needs of tasks with different priorities and improving the practicality and reliability of the entire control system.

[0032] In a preferred embodiment of the present invention, step 1 involves controlling the robot in the mounted state and its mounted vehicle to move together toward the target point in response to a task trigger; during the movement, if an obstacle is detected ahead, it is determined whether the obstacle can be bypassed by the vehicle, including:

[0033] Step 11: parse the target point location information contained in the task trigger;

[0034] Step 12: Based on the target point location information and the current location information of the robot-vehicle assembly, plan a navigation path for the assembly that allows the vehicle to travel;

[0035] Step 13: According to the navigation path of the combined body, control the trolley to carry the robot to move along the navigation path of the combined body.

[0036] Step 14: When the car moves along the navigation path of the combined vehicle, it detects the environment in front in real time, identifies obstacles and obtains their outlines and position information;

[0037] Step 15: Based on the acquired obstacle contours and location information, current environmental information, and the vehicle's motion parameters, analyze and generate at least one potential detour path;

[0038] Step 16: Determine whether at least one potential detour path simultaneously satisfies the path feasibility and distance threshold conditions; if a detour path that satisfies the conditions exists, then it is determined that the obstacle can be bypassed by the vehicle; if no detour path that satisfies the conditions exists, then it is determined that the obstacle cannot be bypassed by the vehicle.

[0039] In this embodiment, by accurately analyzing target point information, combining the current location to plan the vehicle's adaptive path, and moving according to the path, the navigation direction of the combined vehicle is ensured to be accurate and the movement trajectory is compliant, avoiding ineffective driving and improving overall movement efficiency. The environment is detected in real time and obstacle outlines and location information are obtained, providing comprehensive data support for detour judgment, reducing the risk of missed or misjudged obstacles, and ensuring the accuracy of obstacle avoidance decisions. Potential detour paths are generated by combining multi-dimensional information, and detour feasibility is judged by dual verification of path feasibility and distance threshold, which avoids the safety hazards caused by the vehicle forcibly detouring, and also prevents the detour distance from increasing the task time, achieving a balance between scientific obstacle avoidance and efficiency.

[0040] In a preferred embodiment of the present invention, step 11, parsing the target point location information contained in the task trigger, includes:

[0041] Step 1101: Receive a task trigger signal from the task management system;

[0042] Step 1102: Extract the target point identifier or original coordinate data from the task trigger signal;

[0043] Step 1103: Parse and convert the extracted target point identifiers or raw coordinate data into position information in a standard format that the robot navigation system can recognize.

[0044] In this embodiment, receiving a task trigger signal and identifying the signal source as the task management system is the initiation step of the entire target point information parsing process, ensuring that subsequent operations have a legal and valid task trigger basis. From the received task trigger signals, core information related to the target point is filtered out, either the target point identifier (such as the code of a preset point) or the original coordinate data (such as latitude and longitude, local coordinate system coordinates), completing the initial extraction and filtering of information. The extracted target point information is then processed for format standardization, mapping the identifier to the corresponding specific location data, or converting the original coordinate data into a standard format that the robot navigation system can directly recognize and process, achieving information adaptation and connection between the task management system and the navigation system. Only trigger signals from the task management system are received to avoid invalid or illegal signal triggering operations, improve process security, focus on the core information of the target point, reduce redundant data interference, and solve the problem of information format differences between the task management system and the navigation system through format standardization, avoiding navigation failures caused by format incompatibility.

[0045] In a preferred embodiment of the present invention, step 12, based on the target point location information and the current location information of the robot-vehicle assembly, plans a navigation path for the assembly that allows the vehicle to travel, including:

[0046] Step 1201: Obtain the real-time current position information of the robot-vehicle assembly provided by the assembly positioning system;

[0047] Step 1202: Combine the map information provided by the environmental perception module with real-time obstacle information to construct or update the map of the currently passable area;

[0048] Step 1203: Starting from the real-time current position of the composite body and ending with the parsed target point position information, calculate one or more candidate paths on the map of the currently passable area using a path search algorithm.

[0049] Step 1204: Based on the vehicle's physical dimensions, kinematic constraints, and road condition attributes of the candidate paths, select or optimize one path from one or more candidate paths that meets the vehicle's driving capabilities, and use it as the combined navigation path.

[0050] In this embodiment, the real-time current location information is obtained through the combined body positioning system. Compared with methods that rely on a fixed initial position or coarse positioning, this design can dynamically capture the true position of the combined body (such as positional deviations caused by initial placement deviations, residual errors from previous movements, etc.). Accurate starting point information is the core prerequisite for path planning, which can effectively avoid overall path deviations caused by inaccurate starting point positioning, ensuring that the subsequently planned path fundamentally meets the core requirement of traveling from the current position to the target point. By combining map information and real-time obstacle information to construct / update a passable area map, the integration and adaptation of static maps and dynamic environments is achieved. On the one hand, map information provides the basic framework of the environment (such as road boundaries, preset passages, etc.), ensuring that the path conforms to the basic traffic rules of the environment; on the other hand, real-time obstacle information (such as temporarily appearing debris, dynamic obstacles, etc.) can update the passable area in real time, avoiding the inclusion of already occupied areas in the path planning scope. This design ensures that the planned path always conforms to the current real-world environment, effectively reducing the risk of collisions with obstacles during the vehicle's movement and improving safety. It forms a logical closed loop of multi-candidate path generation and targeted selection and optimization, guaranteeing both flexibility in path selection and a good match between the final path and the vehicle's driving capabilities. Generating multiple candidate paths through a path search algorithm avoids the problem of local optima rather than global optima that may exist in single-path planning (e.g., some paths are feasible but too long or have poor road conditions), providing more optimization space for subsequent selection.

[0051] In a preferred embodiment of the present invention, step 14, when the vehicle moves according to the navigation path of the assembly, involves real-time detection of the environment ahead, identification of obstacles, and acquisition of their outlines and position information, including:

[0052] Step 1401: Continuously collect perception data of the area in front by using environmental perception sensors configured on the robot-vehicle assembly;

[0053] Step 1402: Process and analyze the collected perception data to identify one or more obstacles.

[0054] Step 1403: For each identified obstacle, determine its position and contour information in the navigation path coordinate system of the combined body based on the perception data.

[0055] In this embodiment, the environmental perception sensors mounted on the assembly continuously collect perception data of the area ahead, ensuring the real-time and continuous nature of data collection and avoiding missed obstacle detection due to intermittent collection (such as dynamically appearing obstacles not being captured). This provides comprehensive and uninterrupted raw data support for subsequent obstacle identification, ensuring timely obstacle avoidance decisions. The collected perception data is processed and analyzed to identify obstacles. Through implicit processing logic such as data denoising and feature extraction (meeting the requirements of pure text description), invalid interference data (such as changes in light and shadow, differences in ground texture, etc.) is filtered out, improving the accuracy of obstacle identification and reducing the risk of misjudgment (such as misjudging non-obstacles as obstacles) or missed judgment. This ensures that subsequent processing is only performed on real obstacles, avoiding ineffective obstacle avoidance operations. The position and contour information of obstacles are determined in the coordinate system of the assembly's navigation path, achieving coordinate unification between obstacle data and navigation path, avoiding positional deviations caused by inconsistencies in coordinate systems. This provides accurate obstacle spatial parameters for the generation of subsequent potential detour paths, ensuring that detour path planning can accurately avoid obstacles while conforming to the assembly's navigation trajectory, improving the accuracy of obstacle avoidance actions.

[0056] In a preferred embodiment of the present invention, step 15, based on the acquired obstacle contour and location information, current environmental information, and the vehicle's motion parameters, analyzes and generates at least one potential detour path, including:

[0057] Step 1501: Based on the position and outline information of the obstacle, determine a starting point for initiating detour on the navigation path of the combined body;

[0058] Step 1502: Combining the information about the passable area around the obstacle in the current environmental information and the motion parameters of the vehicle, start from the starting point and plan a temporary local path segment that bypasses the outline of the obstacle;

[0059] Step 1503: Connect the temporary local path segment with the original combined navigation path that is not blocked in front of the obstacle to generate a complete potential detour path.

[0060] In this embodiment, the detour starting point is determined based on the precise location and contour information of the obstacle. This avoids problems caused by setting the starting point too early or too late. Detouring too early increases the extra travel distance and reduces movement efficiency, while detouring too late may result in being too close to the obstacle to complete the avoidance maneuver. This ensures that the timing and location of the detour start are precisely adapted to the obstacle shape and the trajectory of the combined vehicle, providing a clear starting point benchmark for subsequent local path planning and improving the rationality of the detour path. By combining the information on the passable area around the obstacle with the vehicle's motion parameters (such as turning radius, travel width, maximum speed, etc.), temporary local path segments are planned, ensuring that the local path conforms to the obstacle. By analyzing the real-world environment surrounding the object and avoiding impassable areas, the path is designed to match the vehicle's physical capabilities. This avoids planning turns or passages that the vehicle cannot perform (such as paths with curvature exceeding the vehicle's maximum turning radius), thus improving the feasibility of local detours. The temporary local path segments are seamlessly integrated with the original combined navigation path, preventing the combined vehicle from deviating from the overall navigation target after detours. This ensures the integrity and continuity of potential detour paths, achieving the core objective of obstacle avoidance while allowing the combined vehicle to quickly return to the main navigation trajectory after detours without needing to replan the entire path, reducing redundant path planning operations and guaranteeing overall movement efficiency.

[0061] In a preferred embodiment of the present invention, if the determination is yes, the vehicle is controlled to bypass the obstacle and continue moving towards the target point; if the determination is no, the vehicle is controlled to stop at a preset convenient location, and the convenient location is recorded as the disembarkation location. The robot is then controlled to disembark and walk across the obstacle to the target point to perform the task, including:

[0062] Step 1A: If it is determined that the obstacle can be bypassed by the vehicle, then proceed with steps 1A1 to 1A3:

[0063] Step 1A1: Select an optimal detour path from at least one potential detour path that meets the conditions;

[0064] Step 1A2: Control the trolley carrying the robot to travel along the final detour path to bypass the obstacles;

[0065] Step 1A3: After successfully bypassing the obstacle, control the car to reconnect to the original or updated combined navigation path and continue moving towards the target point;

[0066] Step 1B: If it is determined that the obstacle cannot be bypassed by the vehicle, then proceed with steps 1B1 to 1B5:

[0067] Step 1B1: Control the car to stop moving forward, and based on the current environmental information, find a convenient parking location within a preset distance range that meets the parking conditions.

[0068] Step 1B2: Control the car to move to the corresponding position and perform a parking operation, and record the geographical coordinates and feature identifiers as the disembarkation position;

[0069] Step 1B3: Control the robot to perform the dismounting action, so that it separates from the vehicle and enters an independent walking state;

[0070] Step 1B4: Based on the target point location information and the robot's current position, plan the robot's walking obstacle-crossing path;

[0071] Step 1B5: Control the robot to move along the walking obstacle-crossing path, cross obstacles on foot, and head to the target point.

[0072] In this embodiment, the optimal path is selected from potential detour paths that meet certain conditions (such as shortest distance, smoothest road conditions, and highest compatibility with the vehicle's motion parameters). Inefficient or potentially risky paths are avoided, minimizing the impact of detours on overall task time and improving stability during the detour process, ensuring smoother vehicle movement. The vehicle is controlled to travel along the optimal detour path, relying on the precisely planned path to prevent deviations and collisions with obstacles, ensuring the robot-vehicle assembly's safety during the detour phase. This also reduces secondary adjustments due to path deviations, improving detour efficiency. After detouring, the original or updated navigation path is reconnected, ensuring the assembly quickly returns to the predetermined navigation target after obstacle avoidance without needing to replan the entire path, reducing redundant path planning operations and ensuring the continuity of the overall movement trajectory. This avoids navigation deviations caused by detours, further improving task execution efficiency. Through a closed-loop logic of optimal path selection, precise detour execution, and rapid return to the main path, in scenarios where obstacles can be bypassed, it achieves the core objective of avoiding obstacles while minimizing the impact of detours on task progress through path optimization and connection design with the main path. It balances driving safety and navigation continuity, fully leveraging the vehicle's maneuverability, avoiding unnecessary robot dismounting operations, and improving the overall efficiency of the combined movement. Within a preset distance range, it finds convenient parking locations that meet the conditions (such as flat roads, no obstructions, no obstruction to other traffic, and easy robot dismounting and subsequent return), avoiding problems such as difficulty in robot dismounting and inability to accurately locate the remounting point upon return due to improper parking locations. This lays a convenient foundation for subsequent robot dismounting, remounting, and obstacle-crossing tasks.

[0073] Recording the geographic coordinates and feature markers of convenient locations as the disembarkation point provides accurate positioning for the robot to return to the vehicle after completing its task. This prevents the robot from forgetting or misjudging its disembarkation point, ensuring the continuity of the disembarkation-task execution-return process and reducing unnecessary search time. Controlling the robot to separate from the vehicle and enter independent walking mode allows for rapid switching of movement modes, fully leveraging the robot's ability to overcome obstacles on foot and compensating for the vehicle's inability to bypass obstacles. This ensures the task is not interrupted by vehicle access restrictions, providing the prerequisite for subsequent obstacle crossing. Based on the target point and the robot's current position, a walking obstacle-crossing path is planned, adapted to the robot's walking motion parameters (such as stride length, obstacle crossing height, turning flexibility, etc.) to avoid planning paths the robot cannot execute, ensuring the feasibility of obstacle-crossing actions and making the robot's movement trajectory more accurate, reducing ineffective movement. Controlling the robot to move along the walking obstacle-crossing path enables it to cross obstacles that the vehicle cannot bypass on foot, ensuring the robot can successfully reach the target point to execute the task. This breaks through the limitations of single-vehicle movement scenarios, ensuring the task can proceed in an orderly manner even in complex obstacle environments.

[0074] Through a layered design of precise parking point selection, location information recording, mode switching, walking path planning, and obstacle crossing execution, efficient collaboration between the vehicle and the robot is achieved in scenarios where obstacles cannot be bypassed. Scientific point selection and location recording ensure smooth boarding and alighting, while mode switching and walking path planning fully leverage the robot's obstacle crossing capabilities, preventing task failure due to obstacles. This significantly improves the system's adaptability to complex obstacle environments and broadens the scope of task execution scenarios.

[0075] In a preferred embodiment of the present invention, step 2, when the task is completed or a new task is received requiring the robot to return to the other side of the obstacle, controls the robot to move to the recorded disembarkation position, including:

[0076] Step 21: When the robot finishes its current task at the target point, or receives a request to move to a new task point on the other side of the obstacle, a return command is triggered.

[0077] Step 22: In response to the return command, retrieve the previously recorded alighting location information from the storage module;

[0078] Step 23: Starting from the robot's current position and ending at the retrieved disembarkation location information, plan a walking navigation path for the robot.

[0079] Step 24: Control the robot to move along the walking navigation path until it reaches the drop-off position.

[0080] In this embodiment, the triggering scenario of the return command is precisely defined (task completion / new task located on the other side of the obstacle) to avoid meaningless return actions and ensure the relevance and necessity of the return command. By clarifying the triggering conditions, the return process is precisely controlled to start only when a return is required, avoiding task interruption or resource waste caused by the robot accidentally triggering the return command, and ensuring the continuity and logic of task execution.

[0081] The system retrieves previously recorded drop-off location information from the storage module, enabling precise reuse of this information. By leveraging recorded geographic coordinates and feature identifiers, it avoids positioning deviations caused by the robot re-searching and reassessing the boarding point. This provides accurate and reliable endpoint data for subsequent navigation path planning, reduces invalid movements due to missing or incorrect location information, and improves the efficiency of the return process.

[0082] A walking navigation path is planned starting from the robot's current position and ending at the retrieved disembarkation position. The path design is fully adapted to the robot's walking movement characteristics (such as stride length, turning flexibility, obstacle crossing ability, etc.). Compared with the navigation path of the car, this walking path is more in line with the robot's independent movement needs, avoiding problems such as walking jams and getting lost due to path mismatch, ensuring the feasibility and efficiency of the navigation path, and shortening the return time.

[0083] The robot is controlled to move along a planned walking navigation path to its disembarkation point, achieving precise arrival from its current location. Closed-loop control ensures the robot strictly follows the planned path, avoiding deviation from the target direction and guaranteeing that the robot can quickly and accurately reach the candidate boarding point. This lays the foundation for subsequent steps such as determining whether the vehicle is in place, performing boarding actions, or locating the vehicle, avoiding process delays caused by deviations in arrival location.

[0084] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for controlling the logic of a robot disembarking to perform a task, characterized in that, The control method includes: In response to a task trigger, the robot and its mounted vehicle are controlled to move together toward the target point. During the movement, if an obstacle is detected ahead, it is determined whether the vehicle can bypass the obstacle. If the vehicle can bypass the obstacle, it continues to move toward the target point. If the vehicle cannot bypass the obstacle, it stops at a preset convenient location and records the convenient location as the disembarkation location. The robot then disembarks and walks across the obstacle to the target point to perform the task. When the robot completes a task or receives a new task that requires it to return to the other side of an obstacle, it controls the robot to move to the recorded disembarkation position; it then determines whether the vehicle is still in the disembarkation position; if yes, it controls the robot to perform the vehicle mounting action, switches to the mounted state, and continues moving; if no, it further determines whether the current task is an emergency task. If the current task is not urgent, control the robot to search for the car within its field of vision. If it finds the car, it will move on the car and continue moving. If it does not find the car, control the robot to walk to perform the subsequent movement. If the current task is an urgent task, control the robot to abandon the search for the car and proceed on foot to carry out the subsequent movement.

2. The robot dismounting and task execution logic control method according to claim 1, characterized in that, In response to a task trigger, control the robot in its mounted state and its mounted vehicle to move together toward the target point, including: Analyze the target point location information contained in the task trigger; Based on the target point location information and the current location information of the robot-vehicle assembly, plan a navigation path for the assembly that allows the vehicle to move. Based on the navigation path of the combined structure, control the trolley to carry the robot to move along the navigation path of the combined structure.

3. The robot dismounting and task execution logic control method according to claim 2, characterized in that, During movement, if an obstacle is detected ahead, it is determined whether the vehicle can bypass it, including: As the vehicle moves along the navigation path of the assembly, it detects the environment ahead in real time, identifies obstacles, and obtains their outlines and position information. Based on the acquired obstacle contours and location information, current environmental information, and the vehicle's motion parameters, analyze and generate at least one potential detour path; Determine whether at least one potential detour path simultaneously satisfies the conditions of path feasibility and distance threshold; if a detour path that satisfies the conditions exists, then it is determined that the obstacle can be detoured by the vehicle; if no detour path that satisfies the conditions exists, then it is determined that the obstacle cannot be detoured by the vehicle.

4. The robot dismounting and task execution logic control method according to claim 3, characterized in that, Parse the target point location information contained in the task trigger, including: Receive task trigger signals from the task management system; Extract the target point identifier or raw coordinate data from the task trigger signal; The extracted target point identifiers or raw coordinate data are parsed and converted into position information in a standard format that can be recognized by the robot navigation system.

5. The robot dismounting and performing task logic control method according to claim 4, characterized in that, Based on the target point location information and the current location information of the robot-vehicle assembly, a navigation path for the assembly that allows the vehicle to travel is planned, including: Obtain the real-time current position information of the robot-vehicle assembly provided by the assembly positioning system; By combining map information provided by the environmental perception module with real-time obstacle information, a map of the currently passable area is constructed or updated. Starting from the real-time current position of the composite object and ending with the parsed target point location information, a path search algorithm is used to calculate one or more candidate paths on the map of the currently passable area. Based on the vehicle's physical dimensions, kinematic constraints, and road condition attributes of candidate paths, a path that meets the vehicle's driving capabilities is selected or optimized from one or more candidate paths and used as the combined navigation path.

6. The robot dismounting and task execution logic control method according to claim 5, characterized in that, As the vehicle moves along the navigation path of the assembly, it detects the environment ahead in real time, identifies obstacles, and obtains their outlines and position information, including: The robot-vehicle assembly continuously collects perception data of the area in front by using environmental perception sensors configured on the robot-vehicle assembly. The collected sensor data is processed and analyzed to identify one or more obstacles. For each identified obstacle, its position and contour information in the navigation path coordinate system of the combined body are determined based on the perception data.

7. The robot dismounting and task execution logic control method according to claim 6, characterized in that, Based on the acquired obstacle contours and location information, current environmental information, and the vehicle's motion parameters, at least one potential detour path is analyzed and generated, including: Based on the location and outline information of the obstacles, a starting point for initiating detours is determined on the navigation path of the combined vehicle; Based on the information about the passable area around the obstacle in the current environmental information, as well as the motion parameters of the car, a temporary local path segment that bypasses the outline of the obstacle is planned starting from the starting point; Connect the temporary local path segment with the original combined navigation path that is not blocked in front of the obstacle to generate a complete potential detour path.

8. The robot dismounting and task execution logic control method according to claim 7, characterized in that, If the determination is yes, the robot is controlled to bypass the obstacle and continue moving towards the target point; if the determination is no, the robot is controlled to stop at a preset convenient location, and this convenient location is recorded as the disembarkation location. The robot then disembarks and walks across the obstacle to the target point to perform the task, including: Step 1A: If it is determined that the obstacle can be bypassed by the vehicle, then proceed with steps 1A1 to 1A3: Step 1A1: Select an optimal detour path from at least one potential detour path that meets the conditions; Step 1A2: Control the trolley carrying the robot to travel along the final detour path to bypass the obstacles; Step 1A3: After successfully bypassing the obstacle, control the car to reconnect to the original or updated combined navigation path and continue moving towards the target point; Step 1B: If it is determined that the obstacle cannot be bypassed by the vehicle, then proceed with steps 1B1 to 1B5: Step 1B1: Control the car to stop moving forward, and based on the current environmental information, find a convenient parking location within a preset distance range that meets the parking conditions. Step 1B2: Control the car to move to the corresponding position and perform a parking operation, and record the geographical coordinates and feature identifiers as the disembarkation position; Step 1B3: Control the robot to perform the dismounting action, so that it separates from the vehicle and enters an independent walking state; Step 1B4: Based on the target point location information and the robot's current position, plan the robot's walking obstacle-crossing path; Step 1B5: Control the robot to move along the walking obstacle-crossing path, cross obstacles on foot, and head to the target point.

9. The robot dismounting and performing task logic control method according to claim 8, characterized in that, When completing a task or receiving a new task requiring a return to the other side of an obstacle, control the robot to move to the recorded disembarkation position, including: When the robot finishes its current task at the target point, or receives a request to move to a new task point on the other side of an obstacle, a return command is triggered. In response to the return command, retrieve the previously recorded alighting location information from the storage module; Starting from the robot's current location and ending at the retrieved disembarkation location information, plan a walking navigation path for the robot. Control the robot to move along the walking navigation path until it reaches the drop-off point.

10. A robot dismounting and performing a task logic control system, characterized in that, include: The control module is used to respond to task triggering and control the robot in the mounted state and the mounted vehicle to move together towards the target point; During movement, if an obstacle is detected ahead, it is determined whether the vehicle can bypass the obstacle. If the judgment is yes, then control the car to go around the obstacle and continue to move towards the target point; if the judgment is no, control the car to stop at a preset convenient location, record the convenient location as the disembarkation location, control the robot to get off the car and walk across the obstacle to the target point to perform the task. The judgment module is used to control the robot to move to the recorded disembarkation position when it has completed a task or received a new task that requires it to return to the other side of the obstacle; to determine whether the vehicle is still in the disembarkation position; if the determination is yes, the robot is controlled to perform the action of getting on the vehicle, and continues to move after entering the carrying state; if the determination is no, the robot is further determined to determine whether the current task is an emergency task. The execution module is used to control the robot to search for a car within its field of vision if the current task is not urgent. If a car is found, the robot will mount the car and continue moving. If a car is not found, the robot will walk to continue moving. If the current task is urgent, the robot will abandon the search for a car and walk directly to continue moving.

Images