98 results about "Robot motion planning" patented technology

The invention relates to the field of mobile robot motion planning, and discloses a quadruped robot motion planning method and system based on height self-adaption. The method comprises the following steps: constructing 3D and 2D local obstacle maps according to coordinate transformation, and calculating a grid vertical distance through ray casting to generate a 2.5 D local obstacle map; determining an initial control sampling space by combining speed, acceleration and height constraints; discretizing a sampling space and simulation time, deducing a control value sequence and recursively calculating a local trajectory; then calculating a shaft alignment bounding box in a directed bounding box of the robot, screening grid units in the shaft alignment bounding box, comparing the height of a machine body with the allowable height of grids, and rejecting collision trajectories to obtain a feasible trajectory set; and finally selecting an optimal track by the evaluation function, extracting a control value and sending the control value to hardware for execution. According to the method, the 2.5 D local obstacle map containing height information is constructed, the four-dimensional control sampling space is designed, and the robot height self-adaptive adjustment is realized in combination with the collision detection and evaluation function.

The invention discloses a motion planning method for a heterogeneous robot, and relates to the technical field of robots. The method comprises the following steps: step 1, subspace definition and differential homeomorphic mapping establishment; step 2, dynamic situation field construction; step 3, geometric motion generation; and 4, controlling system energy. According to the method, universal adaptation of robot systems with different configurations can be realized, the algorithm reuse rate is greatly improved, the development cost is remarkably reduced, efficient obstacle avoidance can be realized in a dynamic environment, and the method has excellent real-time performance and wide applicability and robustness.

The invention relates to the technical field of robot motion path planning, in particular to a beam end spraying and scabbling robot motion planning system based on vision. According to the method, multi-angle images collected by taking a robot as a carrier are subjected to space fusion to generate a three-dimensional space map, position coordinates and posture information of a beam end in the three-dimensional space map are identified and positioned according to standard contour features of the beam end, and an initial arrival path of a robot base is planned; a base movement path is planned according to all the chiseling position coordinates, and a chiseling operation movement path of the mechanical arm is planned based on the defect area and forbidden area positions of the operation coverage area corresponding to all the chiseling position coordinates, so that single-stop coverage maximization and operation safety are both considered; and meanwhile, by combining the actual positions of the base and the tail end of the mechanical arm in the scabbling process of the robot, the positions of the base and the mechanical arm are dynamically verified, the position precision and operation safety in the whole scabbling process are ensured, and the scabbling quality and consistency are improved.

The invention discloses a robot motion planning method fusing GRU-Attention and TD3-SAC, and aims to solve the problem that motion planning based on a traditional reinforcement learning method is relatively weak in strategy diversity, stability and exploration capability balance. Specifically, it is proposed that (1) a gated loop unit (GRU) and an attention module (Attention Module) are introduced into an actor-commentator network to enhance the representation ability of time sequence features; and (2) on the basis of an Actor-Critic network in which GRU-Attention is introduced, a dual-delay depth deterministic strategy gradient (TD3) and a soft actor-commentator (SAC) algorithm are fused, and through fusion of a promotion exploration mechanism of the TD3 and the SAC algorithm, the exploration ability of the robot to a complex unknown environment is enhanced while the training stability and robustness are kept.

The invention provides a multi-mobile-robot layered reinforcement learning motion planning method and system, and belongs to the technical field of robots and swarm intelligence. According to the method, a double-layer planner is constructed; the upper planner and the lower planner perform action planning by adopting a neural network model, and are trained through reinforcement learning. The upper layer planner generates the speed of the robot in a continuous time space according to the local map information and the physical information of the robot, and the speed is used as the global guiding speed and transmitted to the lower layer planner. And the lower layer planner takes optimal reciprocal obstacle avoidance ORCA observation and robot self observation as lower layer observation, and generates a robot action planning result by taking the global guidance speed provided by the upper layer planner as a target speed. By using the method, the problem of multi-robot motion planning in a real indoor unstructured environment can be solved.

The invention belongs to the technical field of path planning and robot motion planning, and particularly relates to a multi-strategy fusion improved RRT * sampling planning method and device and a storage medium, and the method comprises the steps: constructing a dual-search tree through a starting point and a target point in a three-dimensional working space, and selecting a to-be-expanded search tree through a random mode; halton-Bridge mixed sampling is adopted to generate sampling points, so that the global coverage uniformity is improved, and the narrow channel sampling capability is enhanced; calculating the dynamic step length of the exponential decay based on the local obstacle proportion; adopting a hierarchical guidance expansion strategy, preferentially expanding towards a target, introducing an improved artificial potential field to guide expansion when the expansion fails, and taking the bottom with random expansion; performing RRT * neighborhood father node reselection and reconnection optimization on the new node; greedy pruning, self-adaptive interpolation and cubic B-spline smoothing are carried out on a path after the double trees are connected, and an executable smoothing track is output; according to the scheme, planning efficiency, success rate and path quality are improved.

This invention relates to a rapid robot motion planning method. First, it models example actions using a sequence of trajectory points that characterizes task features. Then, it extracts key points from the example trajectory using a trajectory segmentation method based on a first-order difference metric. Finally, it uses PCA to obtain the motion pattern features of the key points of the example trajectory, determines prediction constraints, and achieves rapid motion prediction under a given robot state by solving a quadratic programming problem under these constraints, thus completing the planning. This invention also provides a rapid robot motion planning system. This invention improves the efficiency of robot motion generation in industrial environments and enhances the robot's adaptability to real-time customized tasks.

This invention relates to the field of robot motion planning technology, and proposes a method and system for obstacle avoidance path planning for a multi-degree-of-freedom robotic arm. The method includes: constructing a resultant force guiding field composed of an attractive potential field and a repulsive potential field; iteratively searching from the starting point to the target point along the gradient direction of the resultant force guiding field to generate a pre-path point sequence; mapping the generated pre-path point sequence to the joint space through inverse kinematics to obtain a candidate configuration sequence; sequentially detecting whether the local path segments between adjacent configurations in the candidate configuration sequence satisfy collision constraints, and determining that adjacent configurations are gaps to be repaired if they do not satisfy the constraints; establishing a bidirectional random expansion tree and adjusting the expansion step size in real time according to the potential energy intensity of the repulsive potential field in the current region; repairing the gaps to be repaired through the interconnection of the bidirectional random expansion tree, and splicing the repaired path segments with the path segments that satisfy the collision constraints to generate the final obstacle avoidance motion path.

The invention discloses a robot motion planning method, device and equipment for a constrained trajectory and a medium, and belongs to the field of industrial robot trajectory planning, and the method comprises the steps: converting the motion constraint of a robot at the current moment into a path parameter space; solving a first movement speed of the robot through a third-order sliding mode controller; performing look-ahead simulation on each second constraint boundary set of the robot in the remaining paths; if the simulation result is that the second movement speed exists, the smaller value of the first movement speed and the second movement speed serves as the movement speed of the robot; if the simulation result is that the second movement speed does not exist, the second movement speed of the last period serves as the movement speed of the robot; and controlling the robot to move for a period duration, and if the robot does not walk to the path end point after the motion is finished, circularly executing all the steps until the robot walks to the path end point. Therefore, by implementing the method, real-time planning of the constrained trajectory can be realized.

The invention discloses a robot path planning method based on an adaptive fuzzy and hybrid strategy, and belongs to the technical field of robot motion planning, and the method comprises the steps: carrying out environment perception, and obtaining environment information; the method comprises the following steps: on the basis of an RRT-Connect algorithm framework, fusing a fuzzy control theory, a hierarchical hybrid expansion mechanism and a dynamic path optimization thought to obtain an improved RRT-Connect algorithm; and an improved RRT-Connect algorithm is utilized, and robot path planning is realized based on environment information. According to the method, the fuzzy control theory is utilized to endow the path planning process with a macroscopic intelligent decision-making capability, and a microscopic accurate execution and fault-tolerant capability is provided through a hierarchical mixed strategy, so that the method is an efficient and robust universal solution; the success rate, efficiency and path quality of path planning of the robot in a complex high-dimensional space are remarkably improved, and more efficient and better collision-free path generation can be achieved.

The application belongs to the field of compound robot motion planning and control, and discloses a compound robot dynamic obstacle avoidance control method based on partition adaptive potential field and acceleration MPC fusion. According to the minimum clearance between the robot and the obstacle, the collision, obstacle avoidance and safety area are divided, a continuous partition scheduling function is designed to construct a weighted artificial potential field, a potential field hybrid force is outputted and an MPC initial value sequence is constructed. A discrete prediction model is established by taking acceleration as the control input, a constraint optimization problem in a prediction time domain N=3 is solved online, input constraints, change rate constraints and obstacle avoidance soft constraints with a relaxation variable are applied, the potential field initial value is used for guiding the solution and triggering the initial value updating, and PID weighted terminal guidance and potential field hybrid force fusion are introduced in the target near field. The application considers the safety of dynamic obstacle avoidance, motion stability and target arrival accuracy in a narrow environment.

The application belongs to the field of robots, in order to solve the problems of motion planning scalability, navigation flexibility and system safety of large-scale robot cluster, the application provides a collision risk awareness and high scalability large-scale cluster robot motion planning method and device, comprising: in the macro stage, using a Gaussian mixture model to represent the macro state of the cluster, and planning the transport trajectory of the Gaussian mixture model through model predictive control; in the micro stage, using an artificial potential field method to assign target positions to the robots, and combining a distributed model predictive control to realize the tracking and dynamic obstacle avoidance of the individual to the transport trajectory of the Gaussian mixture model. The application can effectively improve the motion planning performance of the large-scale cluster robot system in a complex environment. In the application, by adjusting the risk acceptance degree, the distance between the robot cluster and the obstacle can be flexibly adjusted, and the balance between safety and motion efficiency is realized.

The application provides a UAV recovery path adaptive planning method and system based on air-ground cooperation, relates to the technical field of robot motion planning, and comprises the following steps: constructing a three-dimensional grid map of a task environment, acquiring the positions of a UAV and an unmanned vehicle and the maximum flight distance of the UAV; determining the responsibility distribution of the UAV and the unmanned vehicle in a recovery task based on the positions of the UAV and the unmanned vehicle and the maximum flight distance of the UAV, and obtaining an optimal UAV recovery scheme; selecting a corresponding path planning mode according to the optimal UAV recovery scheme, and determining the starting point, the ending point and the dimension of the search space of path planning based on the positions of the UAV and the unmanned vehicle; executing a path planning algorithm according to the selected path planning mode, and generating a final recovery path; and through air-ground cooperative decision and cross-dimension bidirectional joint path planning algorithm, the application realizes safe and efficient recovery of the UAV under the constraint of electric quantity, and significantly improves the task success rate and the system cooperation efficiency.

The invention provides an improved CILQR-based intelligent tracked robot motion planning control method and system, and constructs a motion planning control framework which comprises four core modules including an A * algorithm, an improved discrete point quadratic smoothing algorithm, an improved CILQR algorithm and an LQR controller. According to the framework, a motion planning task from a starting point to a target point can be realized based on an environment map fusing static prior information and dynamic real-time information, continuous curvature, safety and no collision of a planned trajectory are ensured, kinematics constraints of an intelligent tracked robot are met, and a finally generated trajectory can be accurately controlled through an LQR controller.

This disclosure provides a motion planning method, apparatus, computer device, and storage medium for robots, applied to each robot in a robot swarm; each robot maintains its own local factor graph; the local factor graphs maintained by the multiple robots constitute a global factor graph; the method includes: in response to a state update event corresponding to the current robot being triggered, determining a trigger domain, and determining a replanning subnet from the current global factor graph based on the trigger domain; performing message passing processing based on Gaussian belief propagation in the replanning subnet with the goal of reducing the energy change of the local factor graph corresponding to the current robot, to obtain an updated local factor graph corresponding to the current robot; determining the confidence distribution information of the variables corresponding to multiple variable nodes in the local factor graph of the current robot based on the updated local factor graph of the current robot; and determining the motion replanning result of the current robot based on the confidence distribution information.

The application relates to a robot motion planning method based on a greedy algorithm and an improved hybrid A* algorithm, and the method comprises the following steps: determining a traversal order of a wall-climbing robot for traversing each detection area by using the greedy algorithm, specifically comprising the following steps: selecting a detection area as a starting point; selecting a detection area closest to the current detection area and not traversed as a next detection area; adding the selected next detection area to a path and marking the detection area as visited; continuously selecting a detection area closest to the current detection area and not traversed as a next detection area until all detection areas are marked as visited; returning to the starting point; and determining the traversal order; and planning a path between adjacent detection areas by using the improved hybrid A* algorithm according to the traversal order. The method has the beneficial effect that the greedy algorithm and the improved hybrid A* algorithm are used to provide a shortest safe path that meets kinematic constraints and can be tracked for the wall-climbing robot.

This application discloses a robot motion planning method, chip, and robot applied to the bottom of furniture. The robot motion planning method includes: when the robot performs a bow-shaped movement, when the robot scans two target support parts on the bottom of the target furniture using a lidar, a scene trigger area corresponding to the target furniture is set based on the two target support parts. Then, the robot enters the scene trigger area at least twice and walks along the line connecting the two target support parts to prevent getting trapped in the hollow area between the two target support parts. During the process of the robot entering the scene trigger area at least twice, it collides with the two target support parts or moves away from the minimum obstacle avoidance distance. Whenever the robot collides with a target support part or moves away from the minimum obstacle avoidance distance, the robot adjusts its movement direction to walk along the line connecting the two target support parts and maintain the bow-shaped movement.

This invention provides a robot motion planning method, apparatus, device, and medium based on a diffusion model. Addressing the shortcomings of existing motion planning methods, such as high dependence on initial solutions, difficulty in balancing planning efficiency and smoothness, and poor generalization ability of learning models, this method proposes an end-to-end framework integrating prior learning and posterior guidance. After acquiring state data and distributing it according to function, the method uses a diffusion model for multi-step iterative denoising starting from random noise. The core of this invention lies in the following: in each denoising time step of trajectory diffusion, the control points of the current intermediate noisy trajectory are extracted and represented by B-spline low-dimensional parameterization; multi-objective comprehensive cost and gradient are constructed in real time based on multi-source states; and the cost gradient is incorporated as a guiding signal into the posterior sampling to guide the denoising update direction of the network. This invention deeply nests dimensionality reduction and cost guidance within a single iteration, enabling the rapid generation of multimodal smooth motion trajectories for adaptive dynamic obstacles without the need for post-processing smoothing.

The application discloses a rigid-flexible coupling rope traction parallel robot motion planning method, equipment and medium, and belongs to the field of rope motion planning. The method comprises the following steps: step 1, a kinematics model and a statics model of the robot are established; step 2, a performance index is constructed based on the established model; step 3, under the known target pose of an end effector, the optimal target pose is obtained by optimizing and solving the target pose through a particle swarm algorithm using the performance index; step 4, a low-dimensional proxy model is constructed based on the established model and the performance index; step 5, the optimal target pose is used as a whole body motion planning target, a path in a mobile platform space and a matching path in a mechanical arm joint space are generated by using the low-dimensional proxy model and a hierarchical planning algorithm, and a whole body motion path is obtained; and step 6, the whole body motion path is processed by using a minimum impact trajectory optimization algorithm, and a whole body motion trajectory for driving the robot is generated. The application can consider obstacle avoidance safety and mechanical stability, and improve quality and execution reliability.

This invention provides a path planning method for industrial robotic arms based on potential field perception and experience-based adaptation, relating to the fields of robot motion planning and intelligent manufacturing technology. Based on the path planning framework of the RRT-Connect algorithm, this method constructs a bidirectional search tree within the search space. It uses the calculated safety margin of the direction-aware space as the initial environmental safety margin for nodes, employs an artificial potential field lateral deflection direction guidance strategy to determine the expansion direction, and uses a parameter adaptive adjustment method with a clearly bounded mapping relationship to adaptively determine the node expansion parameters based on the environmental safety margin of the parent node. This achieves efficient and stable collision-free path planning in complex assembly environments. This method is applicable to path planning problems for multi-degree-of-freedom industrial robotic arms and can significantly improve path planning efficiency and path quality in complex assembly environments.

The invention provides an action planning method, device, equipment and medium for an anti-fracture and anti-compression experimental robot, and relates to the technical field of robot motion planning, and the method comprises the steps: generating a first action plan based on the initial pose information of a to-be-tested block and a first mark area on an anti-fracture instrument, driving a mechanical arm of the experimental robot to move the to-be-tested block to a first target position preset at the fracture resistance instrument, and driving the mechanical arm of the experimental robot to progressively push the to-be-tested block to a test station of the fracture resistance instrument for fracture resistance test; obtaining pose information of a fault block formed after the fracture resistance test and a second mark area on the pressure resistance instrument; and driving a mechanical arm of the experimental robot to move the fault block to a second target position preset at the pressure resistance instrument based on the pose information of the fault block and a second action plan generated by the second marking area, and driving the mechanical arm of the experimental robot to push the fault block to a test station of the pressure resistance instrument in a preset inclined pose to perform a pressure resistance test, and high-robustness robot automatic operation is realized.

The application belongs to the field of robot motion planning and trajectory optimization, and discloses a mobile manipulator space-time trajectory comprehensive performance optimization method based on improved NSGA-III. The method comprises the following steps: step 1, mobile manipulator initial path generation; step 2, trajectory space-time parameterization representation; step 3, multi-objective function establishment, adding specific task constraints, in order to generate smooth, safe and dynamic constraint trajectories for the mobile manipulator to complete specific tasks; step 4, improved NSGA-II algorithm is used to realize trajectory optimization. The method can realize efficient optimization by directly controlling the road point and time through space-time trajectory representation method, reduce the calculation cost in the optimization process, and overcome the difficulty in seeking the optimal solution caused by nonlinearity by using the intelligent optimization algorithm for the multiple targets of obstacle avoidance, path smoothness, dynamic constraint and time optimization of the high-degree-of-freedom mobile manipulator.

A perception and planning method for automatic ball picking of a tennis ball picking robot, mainly including four parts of visual perception, map state updating, path planning and robot motion planning; the method is based on the camera carried by the ball picking robot, uses computer vision and artificial intelligence technology to identify and locate the coordinates of the tennis ball in the local field of view; combines the historical local field of view perception information, performs space-time fusion and task planning, without the need for prior global scanning mapping or relying on global positioning sensors, realizes the improvement of the ball picking efficiency without using more sensors, and does not miss the ball picking area.

To support an operator so as to enable the operator to select an efficient procedure when a robot performs specific work, more easily than a conventional one.SOLUTION: Pieces of procedure data indicating a plurality of procedures when a robot performs specific work are given to a computer. The pieces of procedure data are given by an operator manually operating the robot, for example. The computer generates a Petri net 7 according to α algorithm on the basis of the pieces of procedure data.SELECTED DRAWING: Figure 7

The invention relates to a method and device for determining an unreachable area of a reachable ball of a robot arm and a medium, and the method comprises the steps: extracting geometric parameters of a connecting rod in an arm kinematics chain and joint limiting parameters of joints, carrying out the random sampling according to the joint limiting parameters and a preset sampling density, and obtaining N groups of joint angle combinations, the sampling efficiency is high, joint angle combinations are uniformly distributed, then N point cloud data of an end effector of an arm kinematics chain are determined according to geometric parameters and N groups of joint angle combinations, so that the point cloud data are also uniformly distributed, the reachable ball determined according to the N point cloud data is high in accuracy, and the accuracy of the reachable ball is improved. According to the method, M verification points are randomly generated in the reachable ball, and inverse kinematics solution is performed on each verification point to obtain the unreachable points which do not meet the preset constraint conditions, so that the unreachable area can be accurately and efficiently determined according to the unreachable points, and data support is provided for robot motion planning, structure optimization and man-machine cooperation safety design.

This invention provides a motion planning method for a robotic arm used in live-line operations based on reinforcement learning. First, a reinforcement learning agent and environment model are constructed. Then, a hierarchical reinforcement learning architecture is implemented and trained using a multi-agent reinforcement learning (MARL) framework, treating each robotic arm as an agent. A centralized training distributed execution (CTDE) algorithm is used for collaborative policy learning. Finally, the trained policy network is deployed to a physical robot system. During online operation, only forward computation of the neural network is required, achieving millisecond-level inference. This invention reduces the motion planning time from seconds in traditional methods to milliseconds. The learned policy possesses the ability to generalize to unseen spatial locations with zero or few samples, effectively improving the overall safety and standardization of the operation.

The invention belongs to the technical field of robots, and discloses a robot operation target state characterization method and system based on a look-ahead diffusion network, and the method comprises the steps: obtaining the initial state information of an operation object and environment; the target pose distribution of the operation object is modeled through a diffusion model, and the target pose comprises a translation component and a rotation component; receiving a natural language instruction, and encoding the natural language instruction into a condition feature to guide a generation process of a diffusion model; a back diffusion process of the diffusion model is executed, candidate target poses conforming to the natural language instruction are generated, physical feasibility optimization is conducted on the candidate target poses, and final target poses meeting physical constraints are screened out; and outputting a final target pose for robot motion planning and control. According to the method, multi-modal target pose distribution is explicitly generated through the Gaussian mixture diffusion model, and high-precision, controllable and collision-free target state representation is realized in combination with language condition guidance and gravity-driven correction mechanisms.

The invention discloses a mechanical arm self-adaptive grabbing method based on geometric constraint and gravity optimization, and relates to the field of robot motion planning, and the mechanical arm self-adaptive grabbing method comprises the steps that three-dimensional point cloud data of a target object is acquired, and a smooth surface area is segmented through geometric feature analysis; candidate grabbing point pairs meeting normal vector parallel constraint and friction cone constraint are screened; generating a grabbing strategy including the target clamping force and the grabbing pose based on the candidate grabbing point pair; taking the grabbing pose as a target, and generating a motion track from the initial pose of the mechanical arm to the grabbing pose and with optimized joint load by introducing a path planning algorithm of a gravitational torque cost function; and the mechanical arm is controlled to move along the movement track, and control modes are automatically switched according to the distance and tail end force feedback till grabbing is completed. The stability, the energy efficiency and the safety of the grabbing operation of the mechanical arm in the unstructured environment are improved through geometric physical constraint screening of grabbing points, gravity optimization planning of a movement track and self-adaptive variable stiffness control.

This invention relates to the field of robotic arm motion planning, specifically a multi-robotic arm motion planning method based on a successful experience derivation and replay mechanism. Addressing the low training efficiency and difficulty in learning more complex strategies in multi-robotic arm systems using deep reinforcement learning methods, this invention proposes a successful experience derivation mechanism. By modifying a successful experience, it simulates the robotic arms completing tasks in a tighter formation, thereby deriving successful experiences for more difficult tasks. These successful experiences are then used for experience replay and training, improving the learning effect of the robotic arms. The method includes the following steps: building a simulation environment; modeling the multi-robotic arm motion planning problem; generating tasks; environmental interaction; experience derivation; experience collection and policy updating. Through this invention, the robotic arm can directly learn successful experiences for more complex tasks from successful experiences of simpler tasks, explore collision boundaries to a greater extent, improve sample diversity and coverage, and shorten learning time.

The invention provides a robot motion planning system in a dynamic environment and a robot, the robot motion planning system comprises a positioning unit, a processing unit and a sensing unit, each unit in the processing unit is used for determining the moving tracks of all dynamic obstacles in the environment and dividing the moving tracks into a plurality of time steps; planning a motion path of the first robot and then dividing the motion path into a plurality of time steps; grouping continuous and collision-free time steps between the first robot and the dynamic obstacle into safety intervals; and the first robot scans the motion path of the second robot and searches whether there is a conflict with the motion path of the second robot in a subsequent safety interval, and if there is no conflict, the first robot is controlled to run to the next safety interval. Therefore, a plurality of robots can be planned to move at the same time in the same plane, the robots are effectively prevented from colliding with obstacles and other moving robots, the total consumed time for all the robots to execute tasks is shortest, and the scheduling efficiency of the robots is remarkably improved.