60 results about "Collision free" patented technology

This invention discloses a parking trajectory evaluation method, comprising: generating multiple candidate obstacle avoidance paths represented by fourth-order Bézier curves; utilizing the convex hull property of Bézier curves, a graded collision detection mechanism is used to screen candidate paths with no collision risk through preliminary axial bounding box judgment and precise judgment using the separating axis theorem; for safe candidate paths, regression gravity cost, ride comfort cost, and comprehensive safety cost are calculated by analyzing the geometric relationship of control points, wherein the regression gravity cost is based on the alignment degree between the control point evaluation at the curve endpoint and the heading and curvature of the reference path, and the ride comfort cost is calculated based on the discrete bending energy model of the control polygon; the comprehensive evaluation cost is obtained by weighted summation of the costs of the above three dimensions, and the path with the minimum cost is selected as the execution trajectory, wherein the weight coefficients are dynamically adjusted according to the width of the lateral passable space and the vehicle speed. This invention reduces the computational power consumption of planning and achieves smooth regression with the memorized path after obstacle avoidance.

The application provides a crane group boom anti-collision early warning method and system based on digital twinning, which comprises the following steps: based on the data collected by a rotary encoder, an angle sensor, an inertial measurement unit, a laser radar and a ship body attitude measuring device, a dynamic three-dimensional digital twinning model is constructed and updated in real time; in the model, real-time path planning is performed on each crane boom, the spatial interference risk with the bulkhead, hatch coaming and other booms is evaluated, and a collision-free optimal action path is generated; the path nodes are dynamically risk evaluated, the minimum distance, relative speed, motion direction angle and ship body swing disturbance are comprehensively considered, and a collision risk score is generated; according to the score, hierarchical intervention control is implemented, the collision is avoided, and the attitude stability of the suspended goods is maintained. The application realizes the safety early warning and intelligent regulation and control of multi-machine cooperative operation under high dynamic sea conditions, and effectively improves the operation efficiency and safety.

The invention discloses a self-adaptive hybrid control grinding method and system for a steel rail welding seam. The method comprises the steps that a welding seam three-dimensional model is obtained through 3D vision, and material removal amount distribution is calculated; intelligent layered path planning is conducted based on the distribution, specifically, process threshold layering is set, section sampling is conducted to generate tool points, collision-free optimization is conducted, an initial polishing path is generated, and expected contact force is matched; in the polishing process, the contact force is collected in real time through a six-dimensional force sensor; feedforward control is carried out on the material removal amount based on visual prediction, feedback compensation is carried out through a fuzzy PID controller on the basis of force deviation fed back by a force sensor, motion parameters of the robot are dynamically adjusted, and double-layer self-adaptive control of visual intelligent planning and force control accurate execution is achieved. The system comprises corresponding modules. The problems that an existing method is poor in adaptability to geometric differences of the welding seams and lacks macroscopic planning are solved, and high-precision, high-efficiency and high-consistency automatic grinding of the steel rail welding seams is achieved.

This invention discloses a robotic multi-needle puncture trajectory planning method for prostate cancer. By acquiring two-dimensional ultrasound images and spatial positioning data of the prostate, the coordinates of three-dimensional feature points of obstacles such as blood vessels and the pubic bone are obtained. The tumor contour is extracted using a combination of manual delineation and an improved target detection algorithm to complete three-dimensional reconstruction of the prostate and tumor, as well as implicit function modeling of obstacles, constructing a personalized three-dimensional surgical environment. This method employs a progressive planning strategy, solving for a collision-free reachable workspace for each needle, optimizing the needle insertion trajectory with the shortest path and minimal trauma as objectives, and establishing a tissue deformation compensation model to correct target point offset. Simultaneously, the implanted puncture needles are used as dynamic obstacle iterative update constraints to achieve cyclical optimization of the multi-needle trajectory. Finally, a six-degree-of-freedom robotic arm equipped with a needle guide device completes precise positioning and puncture assistance, achieving orderly, collision-free multi-needle implantation. This invention effectively improves puncture positioning accuracy and surgical safety, reduces manual intervention, and provides an automated, high-precision trajectory planning solution for prostate particle implantation.

The application belongs to the field of path planning, and particularly relates to a robot formation planning method based on a five-dimensional configuration space. The method comprises the following steps: workspace and configuration space definition: the workspace is a two-dimensional plane for robot movement, and the configuration space is a five-dimensional space containing formation centroid, direction and scaling factor; sampling method: random sampling in the configuration space, and ensuring that the sampling points do not cause collision between robots; distance function: defining the distance between two points in the configuration space, and eliminating the dependence on the number of robots; search: searching in the configuration space using an algorithm, and constructing a collision-free path tree; path generation: converting the path in the configuration space into the path in the robot workspace, and smoothing using a spline curve. The robot formation can pass through a narrow channel, and split and merge when encountering a large obstacle. The application provides a high-efficiency, flexible and collision-free global path planning method for a multi-robot system.

This invention discloses an indoor inspection path construction system based on robot vision recognition, belonging to the field of indoor inspection path technology. It obtains synchronously calibrated multimodal perception data; acquires a robot six-DOF pose sequence and sparse 3D point cloud; initializes a 3D Gaussian volume parameter set and constructs an initial 3D Gaussian sputtering model; obtains an optimized robot six-DOF pose sequence and an optimized 3D Gaussian volume parameter set; obtains a continuous collision probability field; obtains a collision-free optimized inspection path; and completes real-time autonomous inspection in an indoor environment. This invention avoids the accidental deletion of key obstacles during downsampling, improving the mapping integrity and pose estimation stability of slender obstacles in the indoor environment.

The present application relates to a method for automatically generating a reference positioning system (RPS) alignment program for shape measurement of a workpiece, the method comprising the steps of: obtaining a reference target of the workpiece from computer aided design (CAD) model information of the workpiece; generating a coordinate system based on the reference target; determining an alignment of the workpiece; and generating a collision-free workpiece program and / or a collision-free movement path for a measurement probe of a shape measurement system using the coordinate system.

The present application relates to the technical field of coal mine safety production monitoring, and specifically discloses a coal mine hidden danger intelligent identification system and early warning device fusing AI video analysis. The system comprises a video acquisition module, an AI visual analysis and calculation module, a three-dimensional situation generation module, an early warning decision and path planning module, a device control module and a feedback adjustment module. The video acquisition module acquires coal flow surface images in real time, the AI visual analysis and calculation module adopts an optical flow method to track particle motion characteristics and outputs arching probability indexes, the three-dimensional situation generation module converts two-dimensional coordinates into arching point three-dimensional coordinates, the early warning decision and path planning module generates a collision-free motion trajectory, the device control module drives the early warning device to perform an arch breaking action, and the feedback adjustment module dynamically adjusts arch breaking parameters according to updated arching probability indexes after arch breaking. The present application realizes early identification, accurate positioning and adaptive disposal of coal bunker blockage hidden dangers in coal mines, and significantly improves the safety level of underground production.

This application belongs to the field of artificial intelligence technology, and provides a home appliance and its interaction processing method, as well as a service robot. The interaction processing method is executed by the service robot. In response to interaction commands, it acquires tag data from a label attached to the surface of the home appliance. This tag data carries geometric positioning information associated with the home appliance and encoded structured information. Based on the geometric positioning information, it calculates the spatial pose of the service robot's visual sensor relative to the label. It decodes and reconstructs the structured information to obtain the three-dimensional structural data of each component of the home appliance and the kinematic constraint data of the movable parts of the home appliance. This kinematic constraint data is used to limit the motion trajectory and physical travel limits of the movable parts. Based on the spatial pose, it transforms the three-dimensional structural data from the label coordinate system to the robot coordinate system to construct a digital twin model of the home appliance. Based on the digital twin model and the kinematic constraint data, it plans the collision-free motion trajectory of the end effector of the service robot's robotic arm. It controls the service robot's robotic arm to perform physical interaction operations with the movable parts according to the collision-free motion trajectory.

The application provides a robot collision detection method based on a friction model and a momentum observer, comprising: establishing a friction model capable of accurately estimating joint friction torque; secondly, constructing a momentum observer that can be adaptively adjusted according to the motion state, for calculating the external torque representing the collision state; further, setting a dynamic threshold range based on the normal motion data without collision, and finally realizing collision determination by comparing the real-time external torque with the threshold. Through high-precision modeling and adaptive observation mechanism, the technical scheme significantly reduces the false positives and false negatives caused by model errors and noise interference, realizes fast and accurate collision detection in the full speed range, and effectively improves the system safety and robustness in human-machine collaboration scenarios.

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 application discloses a kind of robot navigation methods of combining embedded professional knowledge LLM in restricted platform, it is applied to the differential mobile robot of carrying LiDAR, IMU and with support SIMD instruction set MCU as core, the method will embedded professional knowledge LLM and MCU end SIMD light weight DWA fusion, destination coordinates and DWA parameters are generated by LLM analysis natural language instruction, MCU successively completes continuous trajectory modeling, obstacle point cloud SIMD batch division, batch parallel collision detection, selects no collision candidate speed pair and selects optimal value output, realizes robot autonomous target arrival and whole process real-time obstacle avoidance by closed loop control.The application fully taps MCU computing power, improves the real-time of obstacle avoidance and decision efficiency, reduces hardware deployment cost, realizes the deep fusion of planning and execution, operation threshold is low, scene adaptability is strong, engineering realizable is high, and it is suitable for multi-scene differential mobile robot navigation.

The application discloses a kind of based on three-dimensional voxelization map small unmanned aerial vehicle fine inspection path planning method, system, setting and medium, it is related to unmanned aerial vehicle control technical field, the method includes: S1: three-dimensional voxelization modeling is carried out to inspection object and environment, and safety inflation processing is carried out to obstacle;S2: the original inspection view point set is clustered and clustered, and the view point in clustering cluster is sorted according to S type path order;S3: the three-dimensional A* algorithm of optimization is used to generate collision-free polyline path;S4: the line path generated in step S3 is smoothed using line-of-sight algorithm, removes redundant node, generates smooth flight trajectory;S5: the path point after smoothing is converted from local model coordinates to WGS84 latitude and longitude coordinates using decoupling fitting strategy based on control point, generates the GPS waypoint sequence executable by unmanned aerial vehicle.The method provided by the application can solve the problems of unsafe path planning during inspection, low inspection point traversal efficiency, non-smooth flight path and the like in the prior art.

This invention discloses a multi-robot path planning method in complex environments based on multi-strategy self-learning, belonging to the field of multi-robot path planning technology. This method employs the Spiral Forward Adaptive Multi-Objective Artificial Hummingbird Algorithm (SFA-MOAHA) for path planning, enhancing the algorithm's local exploitation and global exploration capabilities by fusing a spiral approximation strategy and a forward fusion learning strategy. Simultaneously, a deep reinforcement learning mechanism is introduced to adaptively select the optimal search strategy based on population state and environmental feedback during the iteration process. Compared to existing technologies, this invention can efficiently solve multi-robot path planning problems in complex dynamic environments, generating high-quality collision-free cooperative paths. It not only effectively avoids collisions between robots and between robots and static / dynamic obstacles, but also significantly shortens the overall path length and improves task completion efficiency, thereby enhancing the safety and robustness of the multi-robot system.

This invention belongs to the field of intelligent driving technology, specifically disclosing a method, system, and computer-readable storage medium for behavior planning at road ramps. Addressing the challenge of balancing traffic efficiency and safety in existing mixed traffic ramp scenarios for connected autonomous vehicles, the invention first establishes a communication model and a multi-view safety prediction model, generating a multi-view graph structure and quantified values ​​of driving behavior safety risks. Then, the behavior planning problem is constructed as an optimization problem with the objective of minimizing total travel time, constrained by collision-free operation and communication quality, and transformed into a graph Markov decision process. Finally, a graph reinforcement learning network is used to solve the problem, outputting the optimal lane-changing operation. This invention effectively balances traffic efficiency and driving safety while ensuring zero collisions and reliable communication, significantly improving decision convergence speed and driving stability.

A method and system for predicting collision-free poses of a motion system, the method comprising: receiving a 3D virtual environment; receiving a 3D representation of the motion system, hereinafter referred to as a 3D virtual motion system, and a set of 3D poses defined for the 3D virtual motion system; receiving a target task to be performed by the motion system relative to the surrounding environment; receiving a predetermined position within the 3D virtual environment, the predetermined position defining the localization in which the 3D virtual motion system must be placed within the 3D virtual environment; applying a collision-free detection function, hereinafter referred to as a CFD function, trained by a machine learning algorithm to an input dataset including the 3D virtual environment, the target task, the predetermined position, and the set of poses, the CFD function being configured to generate a set of collision-free poses as output that enable the motion system to perform the target task when located at the predetermined position; selecting at least one collision-free pose from the set; and displaying the 3D virtual motion system represented by the selected collision-free pose in the 3D virtual environment.

The application discloses a kind of mobile robot predefined time obstacle avoidance tracking control method and system, method includes: the mobile robot control model with unknown nonlinear dynamics and external disturbance is constructed;Establish artificial potential field function containing minimum safety distance and obstacle avoidance detection range;Unknown nonlinear dynamics in the control model is approximated using fuzzy logic system, and system coordinate transformation is given, on this basis, establish Lyapunov function;Under the framework of backstepping recursion method, combined with dynamic surface control technology, construct predefined time filter;Design the adaptive law of mobile robot and predefined time obstacle avoidance controller.The present application is aimed at the problems such as unconstrained convergence time, difficult to balance obstacle avoidance and tracking of existing method, design predefined time filter and obstacle avoidance controller, with the improved artificial potential field function containing key parameters, dynamically adjust control parameters through adaptive law, realize the predefined time convergence of mobile robot system, collision-free accurate tracking.

This invention is a multi-robot encirclement and collision avoidance control method based on pure orientation information, belonging to the field of multi-robot system cooperative control technology. The method includes: constructing a kinematic model of a nonholonomically constrained multi-robot system and a moving target; implementing an observer for each robot based on pure orientation measurement to estimate its relative position to the moving target; using the relative position information estimated by the observer as state feedback and introducing a collision-free reference trajectory generated by the auxiliary multi-robot system, designing a dynamic controller, and ultimately achieving stable encirclement and global collision avoidance of the nonholonomically constrained robots around the moving target. This invention enables state estimation using only the unit vector of relative orientation and robot velocity without relying on distance measurement, achieving multi-robot collision avoidance encirclement and formation control with a fused collision avoidance mechanism based on pure orientation information, and realizing high-precision stable encirclement of maneuvering targets.

This invention discloses a path planning method based on the soft fusion of virtual chemical fields and reinforcement learning, belonging to the field of robot navigation and artificial intelligence technology. The method includes the following steps: constructing a confined space containing obstacles, setting the initial and target positions of the moving subject; constructing a virtual chemical field that attracts the target and repels the obstacle, calculating the field strength characteristics and evolution trend, and scheduling forward movement, fine-tuning, and large-scale turning to generate guiding actions; constructing a reinforcement learning policy model to generate policy actions; introducing an exponential decay soft fusion mechanism to select either a guiding action or a policy action as the final action with decreasing probability; driving the subject to move to a new position, obtaining a reward according to a reward function, and storing the interaction samples in an experience pool; updating the policy model after the samples reach a threshold, repeating until convergence; and generating a collision-free path from the initial position to the target when obstacles change. This invention achieves efficient adaptive collision-free path planning through the soft fusion of biomimetic guidance and reinforcement learning.

The application provides a model training data generation method and system, which comprises the following steps: determining a static scene based on a static scene dataset; sampling a target region in the static scene to determine a target position; the target position is a collision-free motion trajectory planning end point; determining a free space position by using an initial escape mechanism according to a starting position in the static scene; the free space position is a collision-free motion trajectory planning start point of the escape starting position; and performing active obstacle avoidance planning according to the starting position, the free space position and the target position to generate model training motion planning trajectory data. The application can quickly and efficiently generate high-quality motion planning trajectory data, and provide rich spatial information and operation constraints for the model. This not only reduces the data acquisition cost, but also improves the speed and diversity of data generation, and helps to improve the spatial trajectory understanding ability and task execution efficiency of the robot in a complex environment.

This invention relates to a trajectory planning method for unmanned vehicles (UAVs) suitable for narrow, high-curvature road scenarios. It constructs a motion primitive tree based on the vehicle's kinematics model in a Cartesian coordinate system and searches for motion primitives with low collision constraints and minimal costs. These primitives are then sequentially connected to generate an initial trajectory containing spatiotemporal information. The initial trajectory is then smoothed and optimized to minimize energy and time consumption, resulting in a first optimized trajectory. Obstacle safety distance constraints are introduced onto this trajectory, and continuous collision detection is performed in conjunction with the vehicle's swept volume for further optimization, yielding the final drivable trajectory. This invention addresses narrow, high-curvature, unstructured roads. By simultaneously satisfying vehicle kinematics and body shape constraints and integrating spatiotemporal dimensions to achieve joint lateral and longitudinal motion, it obtains the globally optimal trajectory through a complete solution space and combines swept volume and signed distance field (SDF) gradient optimization to achieve global obstacle avoidance and safe driving.

The application belongs to the technical field of electric power robots. A transformer oil taking docking method and system fusing deep learning and force feedback are proposed. A lightweight multi-task model is used to synchronously analyze images and point clouds, to obtain an oil taking port semantic mask, key points and an initial pose. A voxel map containing forbidden areas is constructed based on the mask, and a collision-free trajectory is planned. Cartesian impedance control is used in the initial contact stage to convert force deviation into pose correction to achieve flexible coarse adjustment. After the force is stable, the system switches to a force-position hybrid closed-loop mode, and the joint amount is dynamically corrected according to the real-time force error to maintain torque stability. Through the deep cooperation of vision and force control, the application realizes full-process adaptive control from macro obstacle avoidance to micro precise docking, effectively avoids rigid collision, and significantly improves the precision and flexibility of oil taking operations in complex environments.

This invention discloses an opening and closing mold device for a ground-rail type refrigerator door foaming production line, relating to the technical field of refrigerator door production equipment. It includes a fixing mechanism and an opening and closing mechanism. The fixing mechanism includes a foaming mold frame mounted on the top of a transfer trolley. The foaming mold frame includes a lower mold frame and an upper mold frame hinged together. The lower mold frame is fixed to the transfer trolley, and a hook is fixedly installed on the side wall of the upper mold frame. A locking hook is provided below the hook, and the middle section of the locking hook is rotatably mounted on the lower mold frame. Both the top end of the locking hook and the bottom end of the hook are hook-shaped. A second elastic telescopic rod is provided on the transfer trolley for resetting the locking hook. This opening and closing mold device for a ground-rail type refrigerator door foaming production line achieves collision-free engagement of the foaming mold frame through a cooperative structure of bent plates, bent strips, connecting frames, and rollers. It also provides guidance and limits for the opening of the upper mold frame, preventing free fall or rigid impact closure, protecting the foaming mold, and improving production efficiency.

The application discloses a TBM supporting shoe self-adaptive step changing device, a control system and a method, relates to the technical field of tunnel boring equipment automation, and comprises a data acquisition module (a laser radar array), a data processing module (point cloud denoising and feature extraction) and a control module (path planning and hydraulic servo control). The method realizes real-time scanning of the surrounding rock surface through the laser radar, identifies the steel arch and the collapse cavity features in combination with the improved RANSAC algorithm and PointNet++, automatically plans a collision-free step changing path according to the matching conditions, and precisely controls the displacement of the oil cylinder through the PID. The problems of low positioning accuracy of traditional manual operation, lagging response of the collapse cavity and limited efficiency are solved, and the safety and efficiency of the continuous tunneling of the TBM are remarkably improved.

This invention relates to a method for detecting circuit breaker tripping collisions. A signal triggering device is installed on a moving support. The position of the signal triggering device is adjusted so that when the internal connector of the circuit breaker collides with the moving contact, the top cap contacts the pressure cylinder. An end cover terminal is installed at the circuit breaker flange end cover. The inside of the end cover terminal is electrically connected to a spring via a trigger wire. The break line of a mechanical characteristic meter is electrically connected to the outside of the end cover terminal, and the grounding wire of the mechanical characteristic meter is electrically connected to the moving support terminal. During tripping, if the circuit formed by the break line, end cover terminal, trigger wire, spring, top cap, pressure cylinder, and moving support terminal is continuous, the mechanical characteristic meter displays a break, indicating a tripping collision. If the circuit is not continuous, the mechanical characteristic meter does not display a break, indicating no tripping collision. This detection method eliminates the need for disassembly and reassembly of the circuit breaker, significantly reducing workload and testing cycle.

This invention belongs to the field of power robot technology. It proposes a transformer oil sampling docking method and system integrating deep learning and force feedback. A lightweight multi-task model is used to simultaneously analyze images and point clouds to obtain the semantic mask, key points, and initial pose of the oil sampling port. Based on the mask, a voxel map containing restricted areas is constructed, and a collision-free trajectory is planned. Cartesian impedance control is used in the initial contact phase to convert force deviation into pose correction for flexible coarse adjustment. After the force stabilizes, a force-position hybrid closed-loop mode is switched, dynamically correcting joint parameters based on real-time force errors to maintain torque stability. This invention achieves fully adaptive control from macroscopic obstacle avoidance to microscopic precision docking through deep collaboration between vision and force control, effectively avoiding rigid collisions and significantly improving the accuracy and compliance of oil sampling operations in complex environments.

The utility model provides a kind of complex stiff beam column joint of no-collision insertion construction, including node stirrup, connecting plate, column inner section steel;The node stirrup is arranged on the outside of stiff column longitudinal reinforcement, when the intersection section at beam column joint is larger, two rows of longitudinal reinforcement are arranged in the inside of stiff column longitudinal reinforcement, when beam column is oblique, oblique beam longitudinal reinforcement is connected to connecting plate by reinforcing bar adapter, and connecting plate is fixedly connected with the outside of column inner section steel, the outside surface of connecting plate is inclinedly arranged;The utility model realizes high-quality connection of beam longitudinal reinforcement and column inner section steel at large-section beam column joint, ensures the compactness and durability of node, and provides reliable guarantee for no-collision insertion construction.