A patrol robot active protection method and system

Abstract

The application provides a kind of patrol robot active protection method and system, the method comprises: real-time acquisition of power distribution room scene point cloud and image data, construct global environment grid map and local dynamic obstacle map, while real-time acquisition of patrol robot mobile platform pose data, motion parameters, and the angle, torque and end pose data of each joint of mechanical arm, calculate and generate the multi-level safety protection domain of mobile platform, construct the full-link collision detection safety domain of mechanical arm, and establish the coupling mapping relationship of mobile platform multi-level safety protection domain and mechanical arm full-link collision detection safety domain, and based on the coupling mapping relationship, the protection constraint parameters of mobile platform and mechanical arm are updated synchronously, the motion speed of mobile platform and path planning are controlled in real time, while the joint motion of mechanical arm, end operation trajectory is checked and corrected in real time.The application significantly improves the stability and reliability of patrol robot operation.

Description

Technical Field

[0001] This invention relates to the field of power distribution room inspection technology, specifically to an active protection method and system for an inspection robot. Background Technology

[0002] With the rapid development of intelligent operation and maintenance systems for power distribution networks, unmanned inspection and operation of substations has become a core trend in the industry. Substation scenarios are generally characterized by narrow passages, dense distribution cabinet layouts, and high risks associated with live-line work. This requires inspection robots to possess both autonomous navigation capabilities in narrow passages and the precision operation capabilities of robotic arms operating distribution cabinet panels. The operation process covers the entire process of coordinated movement navigation and precise operation, which places stringent requirements on safety protection throughout the entire operation. Reliable active safety protection technologies have become the core foundation for ensuring stable robot operation and avoiding equipment and personal safety risks.

[0003] Currently, safety protection technologies for robots operating in power distribution rooms mostly focus on independent protection designs for single inspection or single operation functions. Especially in the complex conditions of power distribution rooms—with their confined spaces, dynamic obstacles, and intricate equipment panel layouts—existing single obstacle avoidance or simple stopping protection strategies struggle to balance navigation efficiency with operational accuracy. During dynamic operations, there is a high risk of collisions between the moving platform and environmental obstacles, and of accidental contact between the robotic arm and power distribution cabinet panels and operating components. This can lead to anything from minor task interruptions to serious safety accidents such as short circuits in live equipment and equipment damage.

[0004] Therefore, constructing a multi-layered active safety protection technology that covers the entire process of movement and operation and adapts to dynamic working conditions has become an urgent problem to be solved in the field of power distribution room patrol robots. Summary of the Invention

[0005] Based on this, and in response to the shortcomings of the existing technology, the present invention provides an active protection method and system for patrol robots, in order to solve the problem that the existing technology is unable to balance navigation efficiency and operational accuracy, and that task interruption or even safety accidents are prone to occur during dynamic operations.

[0006] To solve the above-mentioned technical problems, the first aspect of the present invention proposes: An active protection method for a patrol robot, wherein the method includes: The system acquires point cloud and image data of the power distribution room operation scene in real time, constructs a global environment grid map and a local dynamic obstacle map, and collects the pose data and motion parameters of the patrol robot's mobile platform, as well as the angle, torque and end pose data of each joint of the robotic arm in real time. Based on the global environment grid map, local dynamic obstacle map and mobile platform pose data, a multi-level safety protection domain for the mobile platform is calculated and generated through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, joint state data and power distribution cabinet equipment model, a full-link collision detection safety domain for the robotic arm is constructed, and a coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm is established. Based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, the graded protection trigger threshold corresponding to the current working stage is matched, and the protection constraint parameters of the mobile platform and the robotic arm are updated synchronously based on the coupling mapping relationship. Based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters, the motion speed and path planning of the mobile platform are constrained and controlled in real time. At the same time, the joint motion and end-effector trajectory of the robotic arm are checked and corrected in real time to complete the active protection control of the entire operation process.

[0007] The beneficial effects of this invention are as follows: By constructing a global environmental grid map and a local dynamic obstacle map of the power distribution room operation scenario, and simultaneously collecting full-state operation data of the mobile platform and the robotic arm, this invention achieves real-time perception of the operation environment and the robot's own state in all dimensions; by establishing a coupled mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm, it breaks the limitation of the independent operation protection of mobile navigation and robotic arm operation in the prior art; by matching the graded protection trigger thresholds of the operation stage and simultaneously updating the dual-end protection constraint parameters, it effectively balances the robot's navigation efficiency and the precision of the robotic arm's fine operation while ensuring the safety protection capability of the entire process; through the collaborative real-time constraint control and trajectory correction of the mobile platform and the robotic arm, it significantly reduces the probability of task interruption caused by collisions and accidental contact during operation, fundamentally avoiding safety accidents such as equipment damage in live operation scenarios, and significantly improving the stability and reliability of the patrol robot operation.

[0008] Furthermore, the step of calculating and generating a multi-level security protection domain for the mobile platform based on a global environment grid map, a local dynamic obstacle map, and the mobile platform pose data, using a preset security boundary algorithm, includes: Based on the global environment grid map and the outline parameters of the mobile platform, a static security boundary for the mobile platform is constructed. By combining the speed and direction of movement of obstacles in the local dynamic obstacle map, the predicted trajectory envelope of the dynamic obstacle is calculated and generated, and the dynamic safety boundary of the mobile platform is updated based on the predicted trajectory envelope. Based on static safety boundaries, dynamic safety boundaries are superimposed, and three levels of safety protection domains are generated according to distance thresholds from near to far: emergency braking domain, deceleration and avoidance domain, and early warning domain.

[0009] Furthermore, the step of constructing the end-to-end collision detection safety domain of the robotic arm based on the DH parameters, joint state data, and power distribution cabinet equipment model includes: Based on the DH parameters of the robotic arm and the real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. Import the power distribution cabinet equipment model, the operating component model, and the mobile platform body model into the preset collision detection space to complete the coordinate registration of each model; Based on the link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in the collision detection space. The full-link collision detection safety domain is generated according to the minimum distance threshold, which includes an emergency stop domain, a trajectory correction domain, and an attitude adjustment domain.

[0010] Furthermore, the step of establishing the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm includes: Using the world coordinate system as a reference, complete the unified coordinate transformation between the mobile platform base coordinate system and the robotic arm base coordinate system; Based on the pose changes of the mobile platform, the position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Establish a one-to-one mapping relationship between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm, and generate a coupling protection constraint table.

[0011] Furthermore, the step of matching the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform, the operating status of the robotic arm, and the updated data of the local dynamic obstacle map includes: The operation process of the patrol robot is divided into stages, namely the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. For each operation stage, based on the coupling protection constraint table, the distance threshold and speed threshold corresponding to the multi-level safety protection domain of the mobile platform, and the minimum distance threshold and torque threshold corresponding to the full-link collision detection safety domain of the robotic arm are configured respectively; Based on the real-time identified work stage, the corresponding configured hierarchical protection trigger threshold is matched and invoked.

[0012] Furthermore, the step of performing real-time constraint control on the movement speed and path planning of the mobile platform based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters includes: Calculate the actual distance between the mobile platform and obstacles in real time and match the corresponding multi-level safety protection domain; When the actual distance falls into the warning indication range, a warning signal is output and the maximum speed of the mobile platform is limited; When the actual distance falls into the deceleration and avoidance zone, the preset navigation path is locally replanned based on the updated data of the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. When the actual distance falls into the emergency braking zone, the emergency stop command of the mobile platform is triggered, cutting off the drive power output.

[0013] Furthermore, the step of real-time interference verification and correction of the joint movements and end-effector trajectory of the robotic arm includes: Real-time calculation of the actual minimum distance between the robotic arm link-level bounding box model and surrounding models, and matching the corresponding full-link collision detection safety domain; When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time to keep the position of the end effector unchanged. When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is corrected in real time based on the artificial potential field method to avoid interference areas. When the actual minimum distance falls into the emergency stop domain, the robotic arm is triggered to stop, locking the current position of each joint.

[0014] The second aspect of the present invention proposes: An active protection system for a patrol robot, wherein the system includes: The data acquisition module is used to acquire point cloud and image data of the power distribution room operation scene in real time, construct a global environmental grid map and a local dynamic obstacle map, and at the same time acquire pose data and motion parameters of the patrol robot's mobile platform, as well as angle, torque and end pose data of each joint of the robotic arm in real time. The module is used to calculate and generate a multi-level safety protection domain for the mobile platform based on the global environment grid map, the local dynamic obstacle map and the pose data of the mobile platform, through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, the joint state data and the power distribution cabinet equipment model, the module constructs the full-link collision detection safety domain of the robotic arm and establishes the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm. The matching module is used to match the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, and to synchronously update the protection constraint parameters of the mobile platform and the robotic arm based on the coupling mapping relationship. The execution module is used to perform real-time constraint control on the movement speed and path planning of the mobile platform based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters. At the same time, it performs real-time interference verification and correction on the joint movement and end-effector trajectory of the robotic arm, thus completing the active protection control of the entire operation process.

[0015] Furthermore, the acquisition module is specifically used for: Based on the global environment grid map and the outline parameters of the mobile platform, a static security boundary for the mobile platform is constructed. By combining the speed and direction of movement of obstacles in the local dynamic obstacle map, the predicted trajectory envelope of the dynamic obstacle is calculated and generated, and the dynamic safety boundary of the mobile platform is updated based on the predicted trajectory envelope. Based on static safety boundaries, dynamic safety boundaries are superimposed, and three levels of safety protection domains are generated according to distance thresholds from near to far: emergency braking domain, deceleration and avoidance domain, and early warning domain.

[0016] Furthermore, the building module is specifically used for: Based on the DH parameters of the robotic arm and the real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. Import the power distribution cabinet equipment model, the operating component model, and the mobile platform body model into the preset collision detection space to complete the coordinate registration of each model; Based on the link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in the collision detection space. The full-link collision detection safety domain is generated according to the minimum distance threshold, which includes an emergency stop domain, a trajectory correction domain, and an attitude adjustment domain.

[0017] Furthermore, the building module is specifically used for: Using the world coordinate system as a reference, complete the unified coordinate transformation between the mobile platform base coordinate system and the robotic arm base coordinate system; Based on the pose changes of the mobile platform, the position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Establish a one-to-one mapping relationship between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm, and generate a coupling protection constraint table.

[0018] Furthermore, the matching module is specifically used for: The operation process of the patrol robot is divided into stages, namely the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. For each operation stage, based on the coupling protection constraint table, the distance threshold and speed threshold corresponding to the multi-level safety protection domain of the mobile platform, and the minimum distance threshold and torque threshold corresponding to the full-link collision detection safety domain of the robotic arm are configured respectively; Based on the real-time identified work stage, the corresponding configured hierarchical protection trigger threshold is matched and invoked.

[0019] Furthermore, the execution module is specifically used for: Calculate the actual distance between the mobile platform and obstacles in real time and match the corresponding multi-level safety protection domain; When the actual distance falls into the warning indication range, a warning signal is output and the maximum speed of the mobile platform is limited; When the actual distance falls into the deceleration and avoidance zone, the preset navigation path is locally replanned based on the updated data of the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. When the actual distance falls into the emergency braking zone, the emergency stop command of the mobile platform is triggered, cutting off the drive power output.

[0020] Furthermore, the execution module is specifically used for: Real-time calculation of the actual minimum distance between the robotic arm link-level bounding box model and surrounding models, and matching the corresponding full-link collision detection safety domain; When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time to keep the position of the end effector unchanged. When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is corrected in real time based on the artificial potential field method to avoid interference areas. When the actual minimum distance falls into the emergency stop domain, the robotic arm is triggered to stop, locking the current position of each joint.

[0021] The third aspect of the present invention proposes: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the active protection method for patrol robots as described above.

[0022] The fourth aspect of the present invention proposes: A readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the active protection method for patrol robots as described above. Attached Figure Description

[0023] Figure 1 A flowchart of the active protection method for patrol robots provided in the first embodiment of the present invention; Figure 2This is a structural block diagram of the active protection system for patrol robots provided in the third embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0025] Please see Figure 1 The first embodiment of this invention provides an active protection method for patrol robots. This method constructs a global environmental grid map and a local dynamic obstacle map for the power distribution room operation scenario, simultaneously collecting full-state operation data of the mobile platform and the robotic arm, achieving real-time, multi-dimensional perception of the operating environment and the robot's own state. By establishing a coupled mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm, it breaks the limitation of independent mobile navigation and robotic arm operation protection in existing technologies. By matching the graded protection trigger thresholds of the operation stage and simultaneously updating the dual-end protection constraint parameters, it effectively balances robot navigation efficiency and robotic arm precision operation while ensuring full-process safety protection capabilities. Through collaborative real-time constraint control and trajectory correction of the mobile platform and the robotic arm, it significantly reduces the probability of task interruption caused by collisions and accidental contact during operation, fundamentally avoiding safety accidents such as equipment damage in live-line operation scenarios, and significantly improving the stability and reliability of patrol robot operations.

[0026] Specifically, the first embodiment of the present invention provides: An active protection method for a patrol robot, wherein the method includes: Step S10: Real-time acquisition of point cloud and image data of the power distribution room operation scene, construction of global environment grid map and local dynamic obstacle map, and real-time acquisition of pose data, motion parameters, angle, torque and end pose data of each joint of the robotic arm. It should be noted that in this embodiment, the construction of dual maps of the working environment and the full-state acquisition of the robot body are carried out simultaneously. On the one hand, point cloud and image data of the power distribution room working scene are acquired in real time through sensors such as LiDAR and depth cameras. The global environment grid map constructed based on the point cloud data completes the digital modeling of static environments such as walls, fixed power distribution cabinets, and passage boundaries, providing a fixed spatial reference for long-distance navigation of the mobile platform. The locally constructed dynamic obstacle map realizes real-time position tracking and motion status updates of dynamic obstacles such as on-site maintenance personnel and temporary tools, making up for the shortcomings of static maps in adapting to dynamic working conditions on site. On the other hand, the pose, speed, acceleration and other motion parameters of the mobile platform, as well as the angle, torque and end effector pose data of each joint of the robotic arm are collected simultaneously. This realizes the synchronous and full-dimensional acquisition of the operating status of the mobile platform and the robotic arm, avoiding the problems of delayed protection response and misalignment of protection reference caused by the asynchronous acquisition of the two types of data in the prior art, and providing reliable data source support for all subsequent protection calculations.

[0027] Step S20: Based on the global environment grid map, the local dynamic obstacle map and the pose data of the mobile platform, a multi-level safety protection domain of the mobile platform is calculated and generated through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, the joint state data and the power distribution cabinet equipment model, the full-link collision detection safety domain of the robotic arm is constructed, and a coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm is established. It should be noted that, based on the perceived data, dedicated hierarchical safety protection spaces are constructed for the mobile platform and the robotic arm respectively. Then, through coupling mapping relationships, the two independent protection systems are integrated into a coordinated whole, fundamentally breaking down the barriers of independent mobile protection and operation protection without linkage mechanisms in existing technologies.

[0028] Step S30: Based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, match the graded protection trigger threshold corresponding to the current working stage, and based on the coupling mapping relationship, synchronously update the protection constraint parameters of the mobile platform and the robotic arm. It should be noted that this step combines the robot's real-time operating status with dynamic obstacle update information to match the graded protection threshold corresponding to the current operation stage, and synchronously updates the dual-end protection constraint parameters based on the coupling mapping relationship. This solves the problem that the fixed protection threshold in the existing technology cannot adapt to the differentiated needs of different operation stages, and achieves a dynamic balance between protection strength and operation requirements.

[0029] Step S40: Based on the matched graded protection trigger threshold and the updated protection constraint parameters, the motion speed and path planning of the mobile platform are constrained and controlled in real time. At the same time, the joint motion and end-effector trajectory of the robotic arm are checked and corrected in real time to complete the active protection control of the entire operation process.

[0030] It should be noted that this step, based on the matched graded protection trigger threshold and the updated protection constraint parameters, simultaneously performs real-time constraint control on the movement speed and path planning of the mobile platform, and performs real-time interference verification and correction on the joint movement and end-effector trajectory of the robotic arm, ultimately completing active protection control covering the entire operation process of robot movement and operation, forming a complete protection closed loop.

[0031] Second Embodiment Furthermore, the step of calculating and generating a multi-level security protection domain for the mobile platform based on a global environment grid map, a local dynamic obstacle map, and the mobile platform pose data, using a preset security boundary algorithm, includes: Based on the global environment grid map and the outline parameters of the mobile platform, a static security boundary for the mobile platform is constructed. By combining the speed and direction of movement of obstacles in the local dynamic obstacle map, the predicted trajectory envelope of the dynamic obstacle is calculated and generated, and the dynamic safety boundary of the mobile platform is updated based on the predicted trajectory envelope. Based on static safety boundaries, dynamic safety boundaries are superimposed, and three levels of safety protection domains are generated according to distance thresholds from near to far: emergency braking domain, deceleration and avoidance domain, and early warning domain.

[0032] It should be noted that this embodiment constructs the static safety boundary of the mobile platform based on the static environmental information fixed in the global environmental grid map and combined with the shape contour parameters of the mobile platform. This boundary defines the absolute inviolable safety red line of the mobile platform in the fixed environment of the power distribution room, clarifies the basic safe driving space of the robot, and avoids the risk of collision between the mobile platform and static obstacles such as walls and fixed cabinets from the root. It is the insurmountable basic protection foundation of the entire mobile protection system.

[0033] By further combining the movement speed and direction of dynamic obstacles updated in real time in the local dynamic obstacle map, a kinematic prediction algorithm is used to calculate and generate the predicted trajectory envelope of the dynamic obstacles. Based on this predicted trajectory envelope, the dynamic safety boundary of the mobile platform is updated in real time. This upgrades the existing technology's delayed response to the current position of the obstacle to a forward prediction of the obstacle's movement trend, effectively solving the industry pain point of robot jamming caused by untimely or excessive avoidance of dynamic obstacles in narrow passages of power distribution rooms.

[0034] Based on the static safety boundary as an insurmountable foundation, and superimposed with the predictive constraint of the dynamic safety boundary, a three-level safety protection domain is generated according to the distance threshold between the mobile platform and the obstacle from near to far. This domain consists of an emergency braking domain, a deceleration and avoidance domain, and a warning prompt domain. By dividing the risk level into gradients, this technology replaces the single parking protection threshold in the existing technology, providing core model support for balancing navigation efficiency and safety protection capabilities.

[0035] Furthermore, the step of constructing the end-to-end collision detection safety domain of the robotic arm based on the DH parameters, joint state data, and power distribution cabinet equipment model includes: Based on the DH parameters of the robotic arm and the real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. Import the power distribution cabinet equipment model, the operating component model, and the mobile platform body model into the preset collision detection space to complete the coordinate registration of each model; Based on the link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in the collision detection space. The full-link collision detection safety domain is generated according to the minimum distance threshold, which includes an emergency stop domain, a trajectory correction domain, and an attitude adjustment domain.

[0036] It should be noted that this embodiment achieves end-to-end, blind-spot-free, and refined collision protection modeling for the robotic arm, from its base to its end effector. This completely solves the problem of blind spots in existing technologies that only focus on the end effector and ignore the risk of accidental contact between intermediate links and surrounding equipment. Specifically, based on the robotic arm's DH parameters and real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. The DH parameters serve as the core foundation for the robotic arm's kinematic modeling, fully defining the core physical parameters such as the length, torsion angle, and joint offset of each link. Combined with the real-time collected joint angle data, the real-time pose of each link in space can be accurately calculated. The link-level bounding box model constructs a minimum bounding box for collision detection for each link and joint of the robotic arm, rather than a simplified single bounding box for the entire robotic arm in traditional technologies. This completely eliminates the collision detection blind spot for intermediate links of the robotic arm and achieves comprehensive collision detection covering the entire structure of the robotic arm.

[0037] The pre-built power distribution cabinet equipment model, operating component model, and mobile platform body model are uniformly imported into the preset collision detection space, and the coordinates of all models are registered in the same coordinate system. All objects that may collide with the robotic arm, including the target power distribution cabinet, the knobs and switches to be operated, and the mobile platform body installed on the robotic arm itself, are all included in the unified collision detection coordinate system. This ensures the consistency of the coordinate reference of all objects during the collision detection process, avoids distance calculation errors caused by coordinate misalignment, and also avoids the risk of self-collision between the robotic arm and its own mobile platform during operation.

[0038] Based on the completed link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in real time within a unified collision detection space. According to the minimum distance threshold, the entire link collision detection safety domain is divided into emergency stop domain, trajectory correction domain, and attitude adjustment domain, forming a gradient protection system that corresponds one-to-one with the three-level safety protection domain of the mobile platform. Through the division of the hierarchical protection domain, the protection goal of ensuring the robotic arm's operational accuracy in low-risk scenarios and absolute safety in high-risk scenarios is achieved, fundamentally solving the core problem that existing technologies cannot balance operational accuracy and operational safety.

[0039] Furthermore, the step of establishing the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm includes: Using the world coordinate system as a reference, complete the unified coordinate transformation between the mobile platform base coordinate system and the robotic arm base coordinate system; Based on the pose changes of the mobile platform, the position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Establish a one-to-one mapping relationship between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm, and generate a coupling protection constraint table.

[0040] It should be noted that this embodiment solves the root cause problem of the inability of two independent protection systems to synchronize and coordinate. Specifically, by using the global coordinate system of the power distribution room as a unified reference, the coordinate transformation between the mobile platform's base coordinate system and the robotic arm's base coordinate system is completed. This solves the root cause problem of inconsistent coordinate references and inaccurate spatial correspondence between the two execution units, which leads to the inability to synchronize and coordinate protection actions. It provides a unified spatial reference for the coupling of the two safety domains and is a prerequisite for realizing dual-end collaborative protection.

[0041] Based on the pose changes collected in real time by the mobile platform, the absolute position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Since the robot arm's base is fixedly installed on the mobile platform, the pose changes of the mobile platform will directly cause the reference position of the robot arm's entire work space to shift. This step ensures that the coordinate reference of the robot arm's collision detection is always consistent with the actual physical position during the movement of the mobile platform through real-time synchronous updates, completely avoiding the problem of robot arm protection reference misalignment and protection failure caused by the pose changes of the mobile platform.

[0042] A one-to-one mapping relationship is established between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm. A coupled protection constraint table is generated, which predefines the protection constraint rules that the robotic arm needs to execute synchronously when different protection levels of the mobile platform are triggered. For example, when the mobile platform triggers the deceleration and avoidance domain protection, the robotic arm synchronously enters the constraint state of the attitude adjustment domain. When the mobile platform triggers the emergency braking domain protection, the robotic arm must synchronously trigger the emergency stop domain protection. Through this mapping relationship, it is ensured that the protection actions of the mobile platform and the robotic arm are always triggered synchronously and executed in coordination, which completely avoids the risk of secondary collision caused by the mobile platform having triggered braking and stopping while the robotic arm is still moving in the prior art. It truly realizes coordinated protection throughout the entire process of movement and operation.

[0043] Furthermore, the step of matching the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform, the operating status of the robotic arm, and the updated data of the local dynamic obstacle map includes: The operation process of the patrol robot is divided into stages, namely the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. For each operation stage, based on the coupling protection constraint table, the distance threshold and speed threshold corresponding to the multi-level safety protection domain of the mobile platform, and the minimum distance threshold and torque threshold corresponding to the full-link collision detection safety domain of the robotic arm are configured respectively; Based on the real-time identified work stage, the corresponding configured hierarchical protection trigger threshold is matched and invoked.

[0044] It should be noted that this embodiment solves the root cause problem of the inability of two independent protection systems to synchronize and coordinate. The complete operation process of the patrol robot is standardized into stages, distinguishing between the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. The operational objectives, risk characteristics, and core requirements of different stages in the entire process are clearly defined. Specifically, the core requirement of the global navigation stage is to improve long-distance passage efficiency; the core requirement of the power distribution cabinet approach stage is to balance positioning accuracy and obstacle avoidance safety; the core requirement of the fine operation stage is to ensure the operational accuracy of the robotic arm and the safety of live-line work; and the core requirement of the operation retreat stage is smooth reset and obstacle avoidance throughout the process. This standardized stage division provides a clear scenario benchmark for configuring differentiated protection strategies, completely solving the pain point of existing technologies that use fixed protection thresholds and cannot adapt to the differentiated needs of different operational stages.

[0045] For each completed operation stage, based on the coupling protection constraint table, distance thresholds and speed thresholds corresponding to the multi-level safety protection domains of the mobile platform, as well as minimum distance thresholds and torque thresholds corresponding to the full-link collision detection safety domain of the robotic arm, are configured respectively. For example, in the global navigation stage, the distance threshold of the warning prompt domain can be appropriately relaxed and the maximum speed limit threshold can be increased to ensure long-distance passage efficiency. In the fine operation stage, the minimum distance threshold of the robotic arm attitude adjustment domain is narrowed and the upper limit of joint torque is reduced to ensure the accuracy and safety of live operation. Through differentiated threshold configuration, the protection strength and the core requirements of the operation stage are accurately matched.

[0046] Based on the robot's real-time collected motion parameters and work status, the current work stage is identified in real time, and the corresponding pre-configured graded protection trigger threshold is matched and called. At the same time, based on the coupling mapping relationship, the protection constraint parameters of the mobile platform and the robotic arm are updated synchronously, realizing the adaptive and smooth switching of protection strategies throughout the entire work process, ensuring that the protection rules always match the core requirements of the current work.

[0047] Furthermore, the step of performing real-time constraint control on the movement speed and path planning of the mobile platform based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters includes: Calculate the actual distance between the mobile platform and obstacles in real time and match the corresponding multi-level safety protection domain; When the actual distance falls into the warning indication range, a warning signal is output and the maximum speed of the mobile platform is limited; When the actual distance falls into the deceleration and avoidance zone, the preset navigation path is locally replanned based on the updated data of the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. When the actual distance falls into the emergency braking zone, the emergency stop command of the mobile platform is triggered, cutting off the drive power output.

[0048] It should be noted that, based on the real-time updated local dynamic obstacle map and the mobile platform's pose data, the actual distance between the mobile platform and surrounding obstacles is calculated in real time, and the corresponding multi-level safety protection domain is matched. This provides an accurate judgment benchmark for triggering subsequent graded protection actions, ensuring that the protection actions accurately correspond to the actual risk level.

[0049] When the actual distance falls into the warning zone, the robot controller outputs a warning signal and limits the maximum speed of the mobile platform. For low-risk scenarios, without interrupting the navigation task or changing the navigation path, the robot's motion inertia is reduced in advance, thus achieving proactive risk prevention and completely avoiding the problem of decreased work efficiency caused by frequent stops in low-risk scenarios in existing technologies.

[0050] When the actual distance falls into the deceleration and avoidance zone, the preset global navigation path is locally replanned based on real-time updated data from the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. For medium-risk scenarios, smooth obstacle avoidance is achieved through local path planning, effectively avoiding collision risks and ensuring the continuity of navigation tasks, avoiding the problem of the robot stopping and restarting due to a single obstacle avoidance. In the fourth sub-step, when the actual distance falls into the emergency braking zone, an emergency stop command is immediately triggered on the mobile platform, directly cutting off the power output of the drive motor. For high-risk scenarios, instantaneous braking is achieved, eliminating collision accidents from the hardware level and ensuring the absolute safety of the electrical equipment in the power distribution room and the robot itself.

[0051] Furthermore, the step of real-time interference verification and correction of the joint movements and end-effector trajectory of the robotic arm includes: Real-time calculation of the actual minimum distance between the robotic arm link-level bounding box model and surrounding models, and matching the corresponding full-link collision detection safety domain; When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time to keep the position of the end effector unchanged. When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is corrected in real time based on the artificial potential field method to avoid interference areas. When the actual minimum distance falls into the emergency stop domain, the robotic arm is triggered to stop, locking the current position of each joint.

[0052] It should be noted that, based on the real-time updated link-level bounding box model, the actual minimum distance between each link of the robotic arm, the end effector and the surrounding registration model is calculated in real time, and the corresponding full-link collision detection safety domain is matched. This provides a full-link, blind-zone-free trigger judgment benchmark for the robotic arm's graded protection actions, ensuring that every part of the robotic arm's structure is included in the protection range.

[0053] When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time based on the inverse kinematics algorithm, and the spatial pose of the end effector remains unchanged. For low-risk scenarios, without affecting the accuracy of the end operation or interrupting the fine operation, the risk of interference is avoided by adjusting the attitude of redundant joints, and the protective action and the operation task are carried out synchronously without affecting the operation efficiency and operation accuracy.

[0054] When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is smoothed in real time based on the artificial potential field method to actively avoid interference areas. For medium-risk scenarios, a collision-free smooth operation trajectory is replanned on the premise of ensuring that the operation task can be completed smoothly, avoiding interruption of the operation task and ensuring the stability of the operation process.

[0055] When the actual minimum distance falls into the emergency stop zone, the robotic arm's emergency stop command is immediately triggered, locking the current position of all joints. In high-risk scenarios, all movement is immediately halted, preventing accidental contact between the robotic arm and live electrical cabinets or operating components that could lead to equipment short circuits, damage, or other safety accidents. Simultaneously, this coordinated action with the emergency braking of the mobile platform completes the closed-loop execution of the entire active protection method, achieving full coverage and highly reliable active safety protection throughout the entire robot movement and operation process.

[0056] Please see Figure 2 The third embodiment of the present invention provides: An active protection system for a patrol robot, wherein the system includes: The data acquisition module is used to acquire point cloud and image data of the power distribution room operation scene in real time, construct a global environmental grid map and a local dynamic obstacle map, and at the same time acquire pose data and motion parameters of the patrol robot's mobile platform, as well as angle, torque and end pose data of each joint of the robotic arm in real time. The module is used to calculate and generate a multi-level safety protection domain for the mobile platform based on the global environment grid map, the local dynamic obstacle map and the pose data of the mobile platform, through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, the joint state data and the power distribution cabinet equipment model, the module constructs the full-link collision detection safety domain of the robotic arm and establishes the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm. The matching module is used to match the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, and to synchronously update the protection constraint parameters of the mobile platform and the robotic arm based on the coupling mapping relationship. The execution module is used to perform real-time constraint control on the movement speed and path planning of the mobile platform based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters. At the same time, it performs real-time interference verification and correction on the joint movement and end-effector trajectory of the robotic arm, thus completing the active protection control of the entire operation process.

[0057] Furthermore, the acquisition module is specifically used for: Based on the global environment grid map and the outline parameters of the mobile platform, a static security boundary for the mobile platform is constructed. By combining the speed and direction of movement of obstacles in the local dynamic obstacle map, the predicted trajectory envelope of the dynamic obstacle is calculated and generated, and the dynamic safety boundary of the mobile platform is updated based on the predicted trajectory envelope. Based on static safety boundaries, dynamic safety boundaries are superimposed, and three levels of safety protection domains are generated according to distance thresholds from near to far: emergency braking domain, deceleration and avoidance domain, and early warning domain.

[0058] Furthermore, the building module is specifically used for: Based on the DH parameters of the robotic arm and the real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. Import the power distribution cabinet equipment model, the operating component model, and the mobile platform body model into the preset collision detection space to complete the coordinate registration of each model; Based on the link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in the collision detection space. The full-link collision detection safety domain is generated according to the minimum distance threshold, which includes an emergency stop domain, a trajectory correction domain, and an attitude adjustment domain.

[0059] Furthermore, the building module is specifically used for: Using the world coordinate system as a reference, complete the unified coordinate transformation between the mobile platform base coordinate system and the robotic arm base coordinate system; Based on the pose changes of the mobile platform, the position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Establish a one-to-one mapping relationship between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm, and generate a coupling protection constraint table.

[0060] Furthermore, the matching module is specifically used for: The operation process of the patrol robot is divided into stages, namely the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. For each operation stage, based on the coupling protection constraint table, the distance threshold and speed threshold corresponding to the multi-level safety protection domain of the mobile platform, and the minimum distance threshold and torque threshold corresponding to the full-link collision detection safety domain of the robotic arm are configured respectively; Based on the real-time identified work stage, the corresponding configured hierarchical protection trigger threshold is matched and invoked.

[0061] Furthermore, the execution module is specifically used for: Calculate the actual distance between the mobile platform and obstacles in real time and match the corresponding multi-level safety protection domain; When the actual distance falls into the warning indication range, a warning signal is output and the maximum speed of the mobile platform is limited; When the actual distance falls into the deceleration and avoidance zone, the preset navigation path is locally replanned based on the updated data of the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. When the actual distance falls into the emergency braking zone, the emergency stop command of the mobile platform is triggered, cutting off the drive power output.

[0062] Furthermore, the execution module is specifically used for: Real-time calculation of the actual minimum distance between the robotic arm link-level bounding box model and surrounding models, and matching the corresponding full-link collision detection safety domain; When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time to keep the position of the end effector unchanged. When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is corrected in real time based on the artificial potential field method to avoid interference areas. When the actual minimum distance falls into the emergency stop domain, the robotic arm is triggered to stop, locking the current position of each joint.

[0063] The fourth embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the active protection method for patrol robots as described above.

[0064] The fifth embodiment of the present invention provides a readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the active protection method for patrol robots as described above.

[0065] In summary, the active protection method and system for patrol robots provided by this invention can achieve full-dimensional and synchronous perception of dynamic environmental changes and robot operation status through the construction of dual maps of the work scene and the synchronous acquisition of the robot's full state, providing accurate and real-time data support for full-process protection. By constructing a multi-level safety protection domain for the mobile platform and a full-link collision detection safety domain for the robotic arm, and establishing a coupling mapping relationship between the two, the technical barriers between mobile protection and operation protection are completely broken, realizing the coordinated linkage of dual-end protection and eliminating collision detection blind spots and secondary collision risks. Through the adaptive matching of work stage division and graded protection thresholds, the protection strategy and work requirements are accurately adapted, effectively balancing the passage efficiency of the robot's long-distance navigation and the precision of the robotic arm's operation in the power distribution cabinet scenario while ensuring work safety. Through the graded graded protection closed-loop execution, it replaces the single parking protection strategy of the existing technology, greatly reducing the probability of task interruption during operation, fundamentally avoiding safety accidents caused by collisions in narrow spaces and accidental contact with live equipment, and significantly improving the stability, reliability and work efficiency of the power distribution room patrol robot.

[0066] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for active protection of a patrol robot, characterized in that, The method includes: The system acquires point cloud and image data of the power distribution room operation scene in real time, constructs a global environment grid map and a local dynamic obstacle map, and collects the pose data and motion parameters of the patrol robot's mobile platform, as well as the angle, torque and end pose data of each joint of the robotic arm in real time. Based on the global environment grid map, local dynamic obstacle map and mobile platform pose data, a multi-level safety protection domain for the mobile platform is calculated and generated through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, joint state data and power distribution cabinet equipment model, a full-link collision detection safety domain for the robotic arm is constructed, and a coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm is established. Based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, the graded protection trigger threshold corresponding to the current working stage is matched, and the protection constraint parameters of the mobile platform and the robotic arm are updated synchronously based on the coupling mapping relationship. Based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters, the motion speed and path planning of the mobile platform are constrained and controlled in real time. At the same time, the joint motion and end-effector trajectory of the robotic arm are checked and corrected in real time to complete the active protection control of the entire operation process.

2. The active protection method for a patrol robot according to claim 1, characterized in that, The step of calculating and generating a multi-level security protection domain for the mobile platform based on a global environment grid map, a local dynamic obstacle map, and the mobile platform's pose data, using a preset security boundary algorithm, includes: Based on the global environment grid map and the outline parameters of the mobile platform, a static security boundary for the mobile platform is constructed. By combining the speed and direction of movement of obstacles in the local dynamic obstacle map, the predicted trajectory envelope of the dynamic obstacle is calculated and generated, and the dynamic safety boundary of the mobile platform is updated based on the predicted trajectory envelope. Based on static safety boundaries, dynamic safety boundaries are superimposed, and three levels of safety protection domains are generated according to distance thresholds from near to far: emergency braking domain, deceleration and avoidance domain, and early warning domain.

3. The active protection method for a patrol robot according to claim 2, characterized in that, The steps for constructing the end-to-end collision detection safety domain of the robotic arm based on its DH parameters, joint state data, and power distribution cabinet equipment model include: Based on the DH parameters of the robotic arm and the real-time collected joint angle data, a link-level bounding box model of the robotic arm is constructed. Import the power distribution cabinet equipment model, the operating component model, and the mobile platform body model into the preset collision detection space to complete the coordinate registration of each model; Based on the link-level bounding box model, the minimum distance between each link of the robotic arm, the end effector, and each registered model is calculated in the collision detection space. The full-link collision detection safety domain is generated according to the minimum distance threshold, which includes an emergency stop domain, a trajectory correction domain, and an attitude adjustment domain.

4. The active protection method for a patrol robot according to claim 1, characterized in that, The steps for establishing the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm include: Using the world coordinate system as a reference, complete the unified coordinate transformation between the mobile platform base coordinate system and the robotic arm base coordinate system; Based on the pose changes of the mobile platform, the position of the robot arm's base coordinate system in the world coordinate system is updated in real time, and the coordinate reference of the robot arm's full-link collision detection safety domain is corrected simultaneously. Establish a one-to-one mapping relationship between the trigger states of each level of safety protection domain of the mobile platform and the constraint parameters of each level of collision detection safety domain of the robotic arm, and generate a coupling protection constraint table.

5. The active protection method for a patrol robot according to claim 4, characterized in that, The step of matching the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform, the operating status of the robotic arm, and the updated data of the local dynamic obstacle map includes: The operation process of the patrol robot is divided into stages, namely the global navigation stage, the power distribution cabinet approach stage, the fine operation stage, and the operation retreat stage. For each operation stage, based on the coupling protection constraint table, the distance threshold and speed threshold corresponding to the multi-level safety protection domain of the mobile platform, and the minimum distance threshold and torque threshold corresponding to the full-link collision detection safety domain of the robotic arm are configured respectively; Based on the real-time identified work stage, the corresponding configured hierarchical protection trigger threshold is matched and invoked.

6. The active protection method for a patrol robot according to claim 5, characterized in that, The steps for real-time constraint control of the mobile platform's motion speed and path planning based on the matched hierarchical protection trigger threshold and updated protection constraint parameters include: Calculate the actual distance between the mobile platform and obstacles in real time and match the corresponding multi-level safety protection domain; When the actual distance falls into the warning indication range, a warning signal is output and the maximum speed of the mobile platform is limited; When the actual distance falls into the deceleration and avoidance zone, the preset navigation path is locally replanned based on the updated data of the local dynamic obstacle map, and the travel speed of the mobile platform is reduced simultaneously. When the actual distance falls into the emergency braking zone, the emergency stop command of the mobile platform is triggered, cutting off the drive power output.

7. The active protection method for a patrol robot according to claim 6, characterized in that, The steps for real-time interference verification and correction of the joint motion and end-effector trajectory of the robotic arm include: Real-time calculation of the actual minimum distance between the robotic arm link-level bounding box model and surrounding models, and matching the corresponding full-link collision detection safety domain; When the actual minimum distance falls into the attitude adjustment domain, the joint attitude of the robotic arm is optimized in real time to keep the position of the end effector unchanged. When the actual minimum distance falls into the trajectory correction domain, the end-effector trajectory of the robotic arm is corrected in real time based on the artificial potential field method to avoid interference areas. When the actual minimum distance falls into the emergency stop domain, the robotic arm is triggered to stop, locking the current position of each joint.

8. An active protection system for a patrol robot, characterized in that, The system includes: The data acquisition module is used to acquire point cloud and image data of the power distribution room operation scene in real time, construct a global environmental grid map and a local dynamic obstacle map, and at the same time acquire pose data and motion parameters of the patrol robot's mobile platform, as well as angle, torque and end pose data of each joint of the robotic arm in real time. The module is used to calculate and generate a multi-level safety protection domain for the mobile platform based on the global environment grid map, the local dynamic obstacle map and the pose data of the mobile platform, through a preset safety boundary algorithm. At the same time, based on the DH parameters of the robotic arm, the joint state data and the power distribution cabinet equipment model, the module constructs the full-link collision detection safety domain of the robotic arm and establishes the coupling mapping relationship between the multi-level safety protection domain of the mobile platform and the full-link collision detection safety domain of the robotic arm. The matching module is used to match the graded protection trigger threshold corresponding to the current operation stage based on the real-time collected motion parameters of the mobile platform and the working status of the robotic arm, combined with the updated data of the local dynamic obstacle map, and to synchronously update the protection constraint parameters of the mobile platform and the robotic arm based on the coupling mapping relationship. The execution module is used to perform real-time constraint control on the movement speed and path planning of the mobile platform based on the matched hierarchical protection trigger threshold and the updated protection constraint parameters. At the same time, it performs real-time interference verification and correction on the joint movement and end-effector trajectory of the robotic arm, thus completing the active protection control of the entire operation process.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the active protection method for patrol robots as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the active protection method for patrol robots as described in any one of claims 1 to 7.

Images