Electro-hydraulic composite dual-arm robot electric arm deadlock state path planning method and system

Abstract

This invention belongs to the field of electro-hydraulic hybrid dual-arm robot control technology, and provides a path planning method and system for deadlock state of electro-hydraulic hybrid dual-arm robot electric arm. First, based on the number of planning attempts and path search time of the robotic arm from the current position to the target position, it is determined whether a deadlock state has occurred. Then, constraints are gradually relaxed and step-by-step escape control is performed. By judging the number of replanning attempts and path search time in multiple directions, a comprehensive deadlock state judgment is achieved, reducing the risk of false detection and missed detection. Through the constraint gradually relaxed and step-by-step escape control strategy, escape control of electro-hydraulic hybrid dual-arm robot in complex high-altitude environments is realized, ensuring the applicability of the escape control method in electro-hydraulic hybrid dual-arm robot, improving the effectiveness of escape control, and solving problems such as forced execution of escape control and the limitations of escape control on the boundaries of the insulating bucket wall, the space occupied by the hydraulic arm, and the layout of wires and fittings.

Description

Technical Field

[0001] This invention belongs to the field of electro-hydraulic hybrid dual-arm robot control technology, and particularly relates to a path planning method and system for deadlock state of the electric arm of an electro-hydraulic hybrid dual-arm robot. Background Technology

[0002] When an electro-hydraulic hybrid dual-arm robot is in operation, if the load exceeds the load capacity of one of the electric arms, the electric arm will be protected, and it will be necessary to escape from the obstacle. Also, the robotic arm may experience a deadlock state due to being surrounded by obstacles in the surrounding environment during the movement, and may be unable to reach the expected position for a period of time.

[0003] In electro-hydraulic hybrid dual-arm robots, the electric arm is the automatic working arm, while the hydraulic arm requires the cooperation of personnel inside the insulated bucket. During operations such as wire connection and stripping, the end tool of the electric arm is physically connected to the wire and clamp. When trying to escape, it is not possible to force the operation in order to avoid damaging the tool or wire. Different deadlock states need to be considered. In some scenarios, manual confirmation is required before escaping. Moreover, the working space inside the high-altitude insulated bucket is limited, and the escape path is restricted by the boundary of the insulated bucket wall, the space occupied by the hydraulic arm, and the layout of the wire and hardware. This makes it impossible to control the electro-hydraulic hybrid dual-arm robot in the deadlock state of the electric arm using traditional path planning methods. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a path planning method and system for the deadlock state of the electric arm in an electro-hydraulic hybrid dual-arm robot. This invention achieves comprehensive deadlock state detection through multi-directional judgment of the number of replanning attempts and path search time, reducing the risk of false detections and missed detections. By implementing a progressively relaxed constraint and progressive escape control strategy, it achieves escape control for the electro-hydraulic hybrid dual-arm robot in complex high-altitude environments, ensuring the applicability of the escape control method in electro-hydraulic hybrid dual-arm robots, improving the effectiveness of escape control, and solving the problems of forced execution of escape control and the limitations imposed by the boundaries of the insulated bucket wall, the space occupied by the hydraulic arm, and the layout of wires and fittings.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: In a first aspect, the present invention provides a path planning method for the deadlock state of the electric arm of an electro-hydraulic hybrid dual-arm robot, comprising: Deadlock detection is performed by determining whether a deadlock has occurred based on the number of path planning attempts and path search time of the robotic arm from its current position to the target position. Constraints are relaxed step by step. When a deadlock occurs, motion constraints are gradually relaxed and the deadlock is reassessed according to the preset constraint type order. If the deadlock state is still identified after gradually relaxing the motion constraints, the original target is replaced with a preset intermediate target according to the preset target type simplification order for escape control. Electric boom deadlock state planning: When only the electric boom is in a deadlock state, collision detection and path planning between the electric boom and the hydraulic boom are realized according to a preset collision model; wherein, the collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

[0006] Furthermore, the determination of the deadlock state includes: if the robotic arm cannot find a feasible path from the current position to the target position after the number of replanning attempts exceeds a preset number, and / or the path search time exceeds a set threshold, a deadlock state is determined to have occurred.

[0007] Furthermore, the conditions for determining the deadlock state include: the time taken for a single path planning exceeds a threshold, the number of consecutive planning failures exceeds a preset number, the length of the planned path is greater than a preset multiple of the straight-line distance, the degree of enclosure of obstacles around the robotic arm, the joint angle is less than a preset angle, and the external force on the end effector is greater than a preset value.

[0008] Furthermore, the gradual relaxation of the constraint type order includes: gradually relaxing the order of safety distance constraint, path smoothness constraint, joint limit constraint, obstacle avoidance range constraint, and end-effector attitude constraint.

[0009] Furthermore, the target types are simplified into backtracking targets, elevation targets, repositioning targets, and yielding targets.

[0010] Furthermore, the step-by-step escape control includes: recording a sequence of trajectory points over a preset time, searching backward along the trajectory from the current position, and determining the first point whose distance from the current point is greater than a preset distance as the backtracking target; if the backtracking target exists and is reachable, then the escape control stops; otherwise, the current height is detected; if the height is less than the median value of the workspace, then the target to be raised is selected; otherwise, the target to be returned to its original position and the target to be moved aside are tried in turn; if all simplified targets fail, then manual intervention is performed.

[0011] Furthermore, during the escape control process, escape control is implemented according to different escape types in different work processes. Specifically: In lead wire connection operations, when the cutter gets stuck in the insulation layer during wire stripping or the wire clamp gets stuck during connection, the reverse movement is first executed to attempt to rotate the cutter to release stress, reduce the speed, and approach again. If it still fails, it is reported for manual handling. In lead wire cutting operations, when the scissors get stuck in the wire and do not cut completely, the scissor torque is checked. If it exceeds the limit, the cutter is reversed and the cutting point position is changed to cut again. If the number of failures exceeds the preset number, it is reported that the cut is not completely cut. In grounding ring installation operations, when the screw gets stuck during screwing in, it is rotated in the reverse direction to release it, and the speed is reduced to screw it in again. If it still fails, the wire clamp is marked as faulty and the wire clamp is replaced. In fault indicator installation operations, when the latch is not fully triggered and the fault indicator is not locked, the pushing force is increased and the trigger spring is moved up and down. If it fails, it is attempted to push upwards first and then back out, using inertia to trigger, and the installation failure is reported for manual confirmation.

[0012] Furthermore, the coarse layer uses an axis-aligned bounding box structure. It compares the spatial positions of the pose points on the current planned path of the electric arm with the axis-aligned bounding box model of the stationary hydraulic arm, comparing whether the minimum or maximum intervals in each axis overlap. If there is no overlap, the path segment is directly determined to be safe; otherwise, it is marked as a potential collision risk. The standard layer uses convex hull decomposition to convert the joint angle configuration of the electric arm at the path points into the poses of the end effector and each link in the world coordinate system. It uses convex hull decomposition to perform intersection detection between the electric arm link and the stationary hydraulic arm. If a real collision is detected, the path node is discarded and pruned in the planning tree. If it is safe but the distance is less than a preset distance, the minimum distance value is recorded as the repulsive potential field weight or constraint condition in the subsequent trajectory smoothing optimization. If it is completely safe, the node is confirmed to be valid. The fine layer uses a triangular mesh model structure. The verification process calls the original triangular mesh models of the hydraulic arm and the electric arm, as well as the surface model of the electric arm link. The minimum distance between the two models in space is determined by collision detection. If the minimum distance is less than a preset safety threshold, a collision risk is determined, and path replanning is triggered.

[0013] Secondly, the present invention also provides a path planning system for the deadlock state of an electro-hydraulic hybrid dual-arm robot arm, comprising: The deadlock state detection module is configured to determine whether a deadlock state has occurred based on the number of times the robotic arm plans the path from the current position to the target position and the path search time. The constraint relaxation module is configured to: when a deadlock occurs, gradually relax motion constraints and re-evaluate the deadlock state according to the preset constraint type order; The step-by-step escape control module is configured to: if the deadlock state is still determined after gradually relaxing the motion constraints, the original target is replaced with a preset intermediate target according to the preset target type simplification order for escape control; The electric boom deadlock state planning module is configured to: when only the electric boom is in a deadlock state, perform collision detection and path planning between the electric boom and the hydraulic boom according to a preset collision model; wherein, the collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

[0014] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot described in the first aspect.

[0015] Fourthly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the steps of the path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot described in the first aspect.

[0016] Fifthly, the present invention also provides a computer program product, the computer program product comprising a computer program, which, when executed by a processor, implements the steps of the path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot described in the first aspect.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention innovatively proposes a path planning method for deadlock states of the electric arm in an electro-hydraulic hybrid dual-arm robot. First, based on the number of planning attempts and path search time of the robotic arm from its current position to the target position, a deadlock state is determined. Then, constraints are gradually relaxed. When a deadlock state occurs, motion constraints are gradually relaxed and the deadlock state is re-evaluated according to a preset constraint type order. Finally, step-by-step escape control is performed. If a deadlock state is still determined after gradually relaxing motion constraints, the original target is replaced with a preset intermediate target according to a preset target type simplification order for escape control. By judging the number of replanning attempts and path search time from multiple directions, a comprehensive deadlock state judgment is achieved, reducing the risk of false detection and missed detection. Through the step-by-step constraint relaxation and step-by-step escape control strategy, escape control of the electro-hydraulic hybrid dual-arm robot in complex high-altitude environments is realized, ensuring the applicability of the escape control method in electro-hydraulic hybrid dual-arm robots, improving the effectiveness of escape control, and solving problems such as the forced execution of escape control and the limitations of escape control by the boundaries of the insulating bucket wall, the space occupied by the hydraulic arm, and the layout of wires and fittings.

[0018] 2. This invention innovatively proposes a path planning method for deadlock state of the electric arm in an electro-hydraulic hybrid dual-arm robot, and develops a deadlock state judgment method. By judging the number of replanning attempts and path search time of the robotic arm in multiple directions, it realizes all-round deadlock state judgment, reduces the risk of missed judgment, improves the accuracy of deadlock judgment for electric and hydraulic arms, and solves the problem of all-round deadlock state judgment for electric and hydraulic arms with different operation modes under multiple operation types such as wire connection and wire stripping.

[0019] 3. This invention innovatively proposes a path planning method for deadlock state of the electric arm in an electro-hydraulic hybrid dual-arm robot, and develops a constraint-gradually relaxed method. By gradually relaxing the constraints of safety distance, path smoothness, joint limit, obstacle avoidance range, and end-effector posture, it realizes all-round judgment of the electric arm and hydraulic arm in multiple operation scenarios, avoids misjudgment caused by a certain constraint being too high, improves the accuracy of deadlock state judgment, and ensures the accuracy of deadlock state judgment of the electric arm and hydraulic arm in complex operation scenarios.

[0020] 4. This invention innovatively proposes a path planning method for the deadlock state of the electric arm of an electro-hydraulic hybrid dual-arm robot and develops a step-by-step escape control strategy. By simplifying the escape sequence of backtracking the target, raising the target, returning to the target, and yielding the target, the invention realizes the escape control of the electro-hydraulic hybrid dual-arm robot arm in complex high-altitude environments, reduces the control difficulty, ensures the escape effect, and solves the problems of the escape path being limited by the boundary of the insulated bucket wall, the space occupied by the hydraulic arm, and the layout of wires and hardware.

[0021] 5. This invention innovatively proposes a path planning method for the deadlock state of the electric arm in an electro-hydraulic hybrid dual-arm robot, and develops an escape control system for different operation types. The escape control is implemented according to different operation processes and escape types. Specifically: In wire connection operations, when the cutter gets stuck in the insulation layer during wire stripping or the wire clamp gets stuck during connection, the system first performs a reverse motion to attempt to rotate the cutter to release stress and reduce speed to approach again. If this still fails, it is reported for manual handling. In wire cutting operations, when the scissors get stuck in the wire and do not completely cut it, the scissor torque is detected. If it exceeds the limit, the cutter is reversed, the cutting point position is changed, and the cutting is repeated. If the number of failures exceeds a preset number, it is reported that the cut is not completely severed. In grounding ring installation operations, during the screw screwing process... When stuck, rotate in the opposite direction to release, reduce the speed and screw back in. If it still fails, mark the clamp as faulty and replace the clamp. During the installation of the fault indicator, if the latch is not fully triggered and the fault indicator is not locked, increase the pushing force and move the trigger spring up and down. If it fails, try pushing upwards first and then backing up, using inertia to trigger, report the installation failure, and confirm manually. Through targeted escape control strategies for docking lead wire operations, disconnecting lead wire operations, installing grounding ring operations, and installing fault indicators, escape control for different operation types and deadlock states is achieved, improving the targeting of escape control, solving problems such as forced execution and damage to tools or wires, and ensuring the application of escape control strategies in electro-hydraulic hybrid dual-arm robots. Attached Figure Description

[0022] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0023] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0026] Example 1: As described in the background section, in electro-hydraulic hybrid dual-arm robots, the electric arm is the automatic operating arm, while the hydraulic arm requires manual operation from an operator inside the insulated bucket. During operations such as wire connection and stripping, the end effector of the electric arm is physically connected to the wire and clamp, requiring careful consideration before attempting to escape. Forced removal is not permissible to avoid damaging the tool or wire; tool release or reverse action must be performed first. In some scenarios, manual confirmation is necessary before escaping. Furthermore, the working space inside the high-altitude insulated bucket is limited, and the escape path is constrained by the bucket wall boundaries (which cannot be exceeded), the space occupied by the hydraulic arm, and the layout of the wires and fittings. This renders current deadlock detection and escape control methods unsuitable for electro-hydraulic hybrid dual-arm robots.

[0027] To address at least one of the aforementioned problems, this embodiment provides a path planning method for the deadlock state of the electric arm in an electro-hydraulic hybrid dual-arm robot, comprising: S1. Deadlock status determination: S1.1 Deadlock state refers to a state in which the hydraulic arm and / or electric arm, after N replanning attempts (N≥3) during motion planning, still cannot find a feasible path from the current position to the target position, or the path search time exceeds a set threshold (e.g., 10 seconds). Specific deadlock state judgment conditions are shown in Table 1: Table 1 Deadlock Detection Criteria

[0028] S1.2 Determination of critical point: Optionally, based on Monte Carlo simulation, the planning success rate under different working conditions is statistically analyzed in the simulation environment; a planning success rate of <70% is used as the calibration point for each deadlock threshold; the threshold is dynamically adjusted in actual operation and updated after 100 planning cycles.

[0029] S2, Relaxing Constraints Strategy: During operations such as wire connection and stripping, the end effector of the electric boom is physically connected to the wire and clamp. When attempting to escape, the following must be considered: do not forcibly pull it out to avoid damaging the tool or wire; release the tool or perform a reverse action first; in some scenarios, manual confirmation is required before attempting to escape. The working space inside the high-altitude insulated bucket is limited, and the escape path is restricted by: the boundaries of the insulated bucket wall (cannot be exceeded); the space occupied by the hydraulic boom; and the layout of the wire and fittings.

[0030] Therefore, in this embodiment, when a deadlock occurs, the motion constraints are gradually relaxed to expand the feasible solution space, while ensuring safety. Specifically, as shown in Table 2: Table 2 Constraint Hierarchy

[0031] Optionally, the MoveIt planner modifies the collision detection distance threshold through PlanningScene; dynamically adjusts the cost function weights: the safety cost weight is reduced, and the path length weight is increased; when relaxing the limits step by step, the limits are relaxed by one level after each failure, up to a maximum of level 5.

[0032] Based on the gradually relaxed motion constraints, the deadlock state is reassessed during the relaxation process. If the deadlock state is still determined after all constraints are relaxed, then proceed to step S3.

[0033] S3. Simplify the objective: Replace the original goal with a more easily achievable intermediate goal to escape the predicament first, and then replan, as shown in Table 3: Table 3 Simplified Target Types

[0034] Optionally, when simplifying target selection, record the trajectory point sequence of the last 5 seconds, search backward along the trajectory from the current position, and find the first point that is more than 200mm away from the current position as the backtracking target; if the backtracking target exists and is reachable, select priority 1; otherwise, check the current height, if the height is less than the median value of the workspace, select the lifting target; otherwise, try the return target and the yield target in turn; if all simplified targets fail, enter manual intervention.

[0035] S4. Multiple attempts: During the escape and control process, a multi-trial strategy can be adopted, as shown in Table 4: Table 4. Multiple Trial Strategy

[0036] Among them, RRTConnect is suitable for most scenarios and has high bidirectional search efficiency; TRRT (Transition-based RRT) can be equipped with random perturbations and is suitable for narrow channels; BKPIECE is a path planning based on bidirectional search and is suitable for complex environments; the planning time gradually increases, which can give the planner more exploration time; the maximum number of attempts can be 6, and the total time is <68 seconds, avoiding infinite loops.

[0037] S5. Getting out of trouble during each work process: S5.1, Connection of lead wires for getting out of trouble: When the cutting tool gets stuck in the insulation layer during wire stripping, or when the wire clamp gets stuck during wire connection, first perform a reverse motion (exit the wire stripping / connection position), try rotating the cutting tool / sleeve to release stress, and reduce the speed to approach again; if it still fails, report it for manual handling.

[0038] S5.2 Lead wire breaking operation procedure: If the scissors get stuck in the wire and do not cut it completely, check the scissor torque. If it exceeds the limit, reverse the blade and retract it. Change the cutting point position (offset ±5mm) and cut again. If it fails 3 times, report that the steel core has not been completely cut.

[0039] S5.3 Grounding ring installation troubleshooting: If the screw gets stuck during screw insertion (thread misalignment), rotate it in the opposite direction for 2 turns to release it; visually re-inspect the thread alignment; try screwing it in again at a lower speed (30 rpm); if it still fails, mark the clamp as faulty and replace the clamp.

[0040] S5.4, Fault indicator installation troubleshooting: If the latch is not fully triggered and the fault indicator is not locked, increase the pushing force to 80N; move it up and down slightly (±5mm) to trigger the spring; if this fails, try pushing it upwards first and then quickly retracting it to trigger it using inertia; report the installation failure and confirm manually.

[0041] In some other embodiments, when only the electric arm is in a deadlock state (the load exceeds the electric arm's load capacity), collision detection and path planning between the electric arm and the hydraulic arm are realized based on the pose of the hydraulic arm in the camera coordinate system and a preset collision model. The collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

[0042] In the initial planning stage of the electric boom, a coarse layer is used for feasibility assessment. After identifying potential paths, the process switches to a standard layer for fine-grained planning. Once planning is complete, a fine-grained layer is used for final collision verification. The coarse layer is an axis-aligned bounding box structure. It compares the spatial positions of the pose points on the current planned path of the electric boom with the axis-aligned bounding box model of the stationary hydraulic boom, comparing whether the minimum or maximum intervals in each axis overlap. If there is no overlap, the path segment is directly determined to be safe; otherwise, it is marked as a potential collision risk.

[0043] The standard layer employs convex hull decomposition to convert the joint angle configuration of the electric arm at the path points into the poses of the end effector and each link in the world coordinate system. Convex hull decomposition is used to perform intersection detection between the electric arm links and the stationary hydraulic arm. If a real collision is detected, the path node is discarded and pruned in the planning tree. If it is safe but the distance is less than the preset distance, the minimum distance value is recorded as the repulsive potential field weight or constraint condition in the subsequent trajectory smoothing optimization. If it is completely safe, the node is confirmed to be valid.

[0044] The refined layer adopts a triangular mesh model structure. The verification process calls the original triangular mesh models of the hydraulic arm and electric arm, as well as the curved surface model of the electric arm linkage. Collision detection determines the minimum distance between the two models in space. If the minimum distance is less than a preset safety threshold, a collision risk is determined, triggering path replanning. The collision model is divided into multiple levels, as shown in Table 5: Table 5 Collision Model

[0045] Example 2: This embodiment provides a path planning system for deadlock state of an electro-hydraulic hybrid dual-arm robot, including: The deadlock state detection module is configured to determine whether a deadlock state has occurred based on the number of times the robotic arm plans the path from the current position to the target position and the path search time. The constraint relaxation module is configured to: when a deadlock occurs, gradually relax motion constraints and re-evaluate the deadlock state according to the preset constraint type order; The step-by-step escape control module is configured to: if the deadlock state is still determined after gradually relaxing the motion constraints, the original target is replaced with a preset intermediate target according to the preset target type simplification order for escape control; The electric boom deadlock state planning module is configured to: when only the electric boom is in a deadlock state, perform collision detection and path planning between the electric boom and the hydraulic boom according to a preset collision model; wherein, the collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

[0046] The working method of the system is the same as the deadlock state path planning method of the electro-hydraulic hybrid dual-arm robot in Embodiment 1, and will not be repeated here.

[0047] Example 3: This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the deadlock state path planning method for the electro-hydraulic hybrid dual-arm robot described in Embodiment 1.

[0048] Example 4: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, it implements the steps of the deadlock state path planning method for the electro-hydraulic hybrid dual-arm robot described in Embodiment 1.

[0049] Example 5: This embodiment provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the deadlock state path planning method for the electro-hydraulic hybrid dual-arm robot described in Embodiment 1.

[0050] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A path planning method for deadlock state of an electro-hydraulic hybrid dual-arm robot arm, characterized in that, include: Deadlock detection is performed by determining whether a deadlock has occurred based on the number of path planning attempts and path search time of the robotic arm from its current position to the target position. Constraints are relaxed step by step. When a deadlock occurs, motion constraints are gradually relaxed and the deadlock is reassessed according to the preset constraint type order. If the deadlock state is still identified after gradually relaxing the motion constraints, the original target is replaced with a preset intermediate target according to the preset target type simplification order for escape control. Electric boom deadlock state planning: When only the electric boom is in a deadlock state, collision detection and path planning between the electric boom and the hydraulic boom are realized according to a preset collision model; wherein, the collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

2. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 1, characterized in that, The deadlock state determination includes: if the robotic arm cannot find a feasible path from the current position to the target position after the number of replanning attempts exceeds a preset number, and / or the path search time exceeds a set threshold, a deadlock state is determined to have occurred.

3. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 2, characterized in that, The deadlock conditions include: the time taken for a single path planning exceeds a threshold, the number of consecutive planning failures exceeds a preset number, the length of the planned path is greater than a preset multiple of the straight-line distance, the degree of enclosure of obstacles around the robotic arm, the joint angle is less than a preset angle, and the external force on the end effector is greater than a preset value.

4. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 1, characterized in that, The gradual relaxation of the constraint types includes the following: the gradual relaxation of safety distance constraints, path smoothness constraints, joint limit constraints, obstacle avoidance range constraints, and end-effector attitude constraints.

5. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 1, characterized in that, The simplified order of the target types is: backtracking target, elevation target, return target, and yielding target.

6. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 5, characterized in that, The step-by-step escape control includes: recording a sequence of trajectory points over a preset time period; searching backward along the trajectory from the current position to determine the first point whose distance from the current point is greater than a preset distance as the backtracking target; if the backtracking target exists and is reachable, then the escape control stops; otherwise, the current height is detected; if the height is less than the median value of the workspace, then the target to be raised is selected; otherwise, the target to be returned to its original position and the target to be moved aside are tried in turn; if all simplified targets fail, then manual intervention is performed.

7. The path planning method for deadlock state of the electro-hydraulic hybrid dual-arm robot as described in claim 1, characterized in that, During the aforementioned escape control, escape control is implemented according to different escape types in different work processes. Specifically: In lead wire connection operations, when the cutter gets stuck in the insulation layer during wire stripping or the clamp gets stuck during connection, the reverse motion is first executed to attempt to rotate the cutter to release stress, reduce the speed, and approach again. If it still fails, it is reported for manual handling. In lead wire cutting operations, when the scissors get stuck in the wire and do not completely cut it, the scissor torque is checked. If it exceeds the limit, the cutter is reversed and the cutting point position is changed to cut again. If the number of failures exceeds the preset number, it is reported that the cut is not completely cut. In grounding ring installation operations, when the screw gets stuck during screwing in, it is rotated in the reverse direction to release it, and the speed is reduced to screw it in again. If it still fails, the clamp is marked as faulty and the clamp is replaced. During the installation of the fault indicator, if the latch is not fully triggered and the fault indicator is not locked, increase the pushing force and move the trigger spring up and down. If this fails, try pushing it upwards first and then backing it out, using inertia to trigger the installation. Report the installation failure and confirm manually.

8. The path planning method for deadlock state of the electric arm of an electro-hydraulic hybrid dual-arm robot as described in claim 1, characterized in that, The coarse layer uses an axis-aligned bounding box structure. It compares the spatial positions of the pose points on the current planned path of the electric arm with the axis-aligned bounding box model of the stationary hydraulic arm, comparing whether the minimum or maximum intervals in each axis overlap. If there is no overlap, the path segment is directly determined to be safe; otherwise, it is marked as a potential collision risk. The standard layer uses convex hull decomposition to convert the joint angle configuration of the electric arm at the path points into the poses of the end effector and each link in the world coordinate system. It uses convex hull decomposition to perform intersection detection between the electric arm links and the stationary hydraulic arm. If a real collision is detected, the path node is discarded and pruned in the planning tree. If it is safe but the distance is less than a preset distance, the minimum distance value is recorded as the repulsive potential field weight or constraint condition in the subsequent trajectory smoothing optimization. If it is completely safe, the node is confirmed to be valid. The fine layer uses a triangular mesh model structure. The verification process calls the original triangular mesh models of the hydraulic arm and the electric arm, as well as the surface model of the electric arm link. Collision detection determines the minimum distance between the two models in space. If the minimum distance is less than a preset safety threshold, a collision risk is determined, and path replanning is triggered.

9. A path planning system for deadlock state of an electro-hydraulic hybrid dual-arm robot arm, characterized in that, include: The deadlock state detection module is configured to determine whether a deadlock state has occurred based on the number of times the robotic arm plans the path from the current position to the target position and the path search time. The constraint relaxation module is configured to: when a deadlock occurs, gradually relax motion constraints and re-evaluate the deadlock state according to the preset constraint type order; The step-by-step escape control module is configured to: if the deadlock state is still determined after gradually relaxing the motion constraints, the original target is replaced with a preset intermediate target according to the preset target type simplification order for escape control; The electric boom deadlock state planning module is configured to: when only the electric boom is in a deadlock state, perform collision detection and path planning between the electric boom and the hydraulic boom according to a preset collision model; wherein, the collision model includes a coarse layer for collision detection, a standard layer for planning, and a fine layer for final path verification.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the deadlock state path planning method for the electro-hydraulic hybrid dual-arm robot as described in any one of claims 1-8.

Images