127 results about "Bidirectional search" patented technology

The invention discloses a robot path planning method and system and a storage medium, and the method comprises the steps: building an environment topology model based on sensor data and map information, building an adaptive map data structure, carrying out the global planning of a rough level through employing a low-resolution map, and carrying out the local optimization of a detail level through employing a high-resolution map. Adaptive dynamic switching is carried out between the two levels according to a path planning task; constructing a two-way alternate search strategy, and performing path search by combining an environment topology model and map information; environment change is monitored in real time, the connectivity probability of each edge in the probability graph is dynamically updated, and a path search strategy is adaptively adjusted according to the change of the probability graph; and establishing a backup path selection model based on reinforcement learning, and dynamically adjusting a backup path in combination with environment change and main path search. Through a bidirectional alternate search strategy, the path search efficiency is improved, the problem of path repetition caused by bidirectional search is solved, and the optimal path from the starting point to the end point can be found more quickly.

The invention discloses an intelligent storage robot path planning method, and belongs to the technical field of intelligent robots, and the method comprises the steps: improving a bidirectional A-Star algorithm, and combining a Time Elasticic-Band (TEB) obstacle avoidance algorithm, and providing a new intelligent storage path planning method. The improved A-Star algorithm adopts bidirectional search to replace unidirectional search of a traditional A-Star algorithm; 5-neighborhood search is adopted to replace 8-neighborhood search; introducing a self-adaptive dynamic search factor to improve a heuristic function; and smoothing the path by adopting a redundant node removal and linear interpolation method. The improved bidirectional A-Star algorithm searches to obtain a global optimal path, and the TEB algorithm optimizes and adjusts a local path according to the global path to complete dynamic obstacle avoidance. According to the method, the speed and precision of path planning are effectively improved, the local path can be adjusted in real time to avoid dynamic obstacles, and the requirement for path planning of the intelligent storage robot is met.

The invention relates to a path planning method based on an improved bidirectional A star search algorithm, relates to the technical field of intelligent vehicles, and solves the problem of low search efficiency caused by bloated search nodes or special topographic reasons in an existing path search scheme. The path planning method based on the improved bidirectional A star search algorithm comprises the following steps: acquiring an obstacle grid map of a target scene; determining a motion path from a starting point position to a target position in the barrier grid map based on an optimized bidirectional A * search algorithm; wherein in the optimized bidirectional A * search algorithm, the forward heuristic function value of the child node is a heuristic function value from the child node to the child node with the minimum reverse total cost at present, and the reverse heuristic function value of the child node is a heuristic function value from the child node to the child node with the minimum forward total cost at present.

The invention relates to the technical field of path planning, and discloses a real-time path data planning processing method and system based on SLAM. The method comprises the following steps: performing semantic-geometric joint feature extraction through a laser radar and an RGB camera to obtain joint observation data; performing covariance propagation on the SLAM state vector based on semantic constraint Kalman filtering to obtain a positioning covariance and a covariance ellipse; performing constraint propagation on the semantic object set to generate virtual feature points to construct an enhanced feature map; environment grid safety scoring is carried out according to the semantic risk weight and the obstacle distance; semantic cost and SLAM uncertainty processing is carried out through bidirectional A * search, and a real-time navigation path is generated. According to the method, the technical problem of insufficient comprehensive consideration of semantic security constraint and positioning uncertainty in real-time path planning based on SLAM in a feature sparse environment is solved.

The invention relates to a multi-AGV path planning method and device based on a time window and computer equipment. The method comprises the following steps: constructing a working environment map of AGVs by adopting a grid method; according to the working environment map, performing path planning on each AGV by adopting an improved Floyd bidirectional search JPS algorithm to obtain an initial global optimal path when each AGV executes a respective delivery task, constructing a time window model of the AGV, and according to the time window model and the initial global optimal paths when all the AGVs execute respective delivery tasks, constructing a time window model of the AGV, and according to the time window model and the initial global optimal paths when all the AGVs execute the respective delivery tasks, constructing a time window model of the AGV; and local path planning is carried out by adopting a conflict resolution strategy in which a translation time window algorithm and a re-planning strategy are parallel, and a conflict-free path of each AGV is obtained. According to the method, aiming at possible conflicts among multiple AGV paths, local path planning is carried out by adopting a time window and replanning mixed conflict resolution method, conflicts among the AGVs are avoided, and more reliable and efficient gains are achieved.

The invention discloses an outdoor robot path planning method based on improved bidirectional A * and DWA algorithms. The method comprises the following steps: step 1, A * algorithm improvement; step 2, DWA algorithm improvement; step 3, algorithm fusion; step 4, path planning; the dynamic weight DOMW strategy is introduced into the A * algorithm to optimize a heuristic function, multiple factors are comprehensively considered, path planning is more reasonable, the algorithm calculation amount and search time are reduced by removing redundant nodes, the algorithm efficiency is improved, a bidirectional search mechanism is adopted, and the algorithm efficiency is improved. The search efficiency is improved, the search space is reduced, a local optimal solution is avoided, the robustness of the algorithm is enhanced, the path is smoothed by adopting a Bezier curve optimization strategy, so that the path meets the actual motion requirement of the robot, the damage to hardware is reduced, and the attractiveness and readability of the path are improved; by optimizing a DWA algorithm trajectory evaluation function and speed space constraints, the collision risk is reduced, and local obstacle avoidance and operation stability are improved.

The invention discloses an intelligent detection robot system for a valve hall assembly of a power converter station and a detection method. A robot body adopts an overhead six-axis series mechanical arm structure and is suspended in the center of the top of a valve hall; the intelligent path planning module performs optimization by combining a probability path diagram method with an improved A * bidirectional search algorithm, and ensures that the robot has efficient obstacle avoidance capability and an optimal moving track in a complex environment; the point cloud data detection and processing module adopts a multi-sensor fusion technology, combines an ICP algorithm to complete point cloud registration, and generates a three-dimensional model with labels. Compared with traditional manual detection, the system has the advantages that the detection efficiency and safety are improved, the labor cost is remarkably reduced, the development of intelligent maintenance means of electric power facilities is promoted, and technical guarantee is provided for stable operation of a converter station.

The invention relates to a mobile robot path planning method and system, and the method comprises the steps: determining a multi-neighborhood mixed search mechanism for the mixed use of eight neighborhoods and new sixteen neighborhoods according to whether there is an obstacle in the eight neighborhoods of a current node; deleting invalid nodes according to the positions of the current node and the target node; a bidirectional search strategy is adopted, a dynamic weight coefficient is added to adjust the heuristic function, and a planned path is obtained; and a redundant inflection point deletion strategy is adopted for the planned path, useless inflection points are removed, and an optimal path is obtained. By improving a neighborhood search mechanism, the search efficiency of a path planning algorithm is improved; through a node optimization strategy, invalid nodes in the algorithm searching process are removed, and the path planning algorithm searching time is shortened; by introducing a two-way search mechanism, the algorithm search efficiency is improved, meanwhile, the dynamic weight factor is added to adjust the algorithm search speed, search of invalid regions is reduced, and therefore the search efficiency is improved.

The invention provides a comprehensive scene map path planning method based on an improved A * algorithm, and the method comprises the steps: 1, carrying out environment modeling for a comprehensive scene map: segmenting a two-dimensional plane scene into more than two square grids, and carrying out the assignment of a single grid in a binarization manner to represent the occupation condition of the grids; constructing a protection area formed by a single-layer grid around the obstacle, and filling the protection area into gray; step 2, planning a path by adopting an improved A * algorithm: step 2-1, establishing an improved bidirectional search strategy; step 2-2, establishing a heuristic function with a barrier linear density weight; step 2-3, adopting a variable neighborhood node expansion mode to complete node expansion; and step 2-4, carrying out path secondary optimization. According to the method, the planning efficiency and the path performability in a complex environment are remarkably improved, and an effective solution is provided for navigation systems in the fields of service robots, intelligent warehousing and the like.

The invention relates to the technical field of unmanned aerial vehicle trajectory planning, in particular to a bidirectional tree random search method based on a complex environment node cost function, which comprises the steps of surveying obstacle data in a flight area and modeling, then introducing a target deviation strategy and a bidirectional search tree mechanism, and combining with an improved minimum cost method to obtain a random search result. Alternately expanding the two trees to generate an initial path, and integrating the path length, the obstacle distance and the flight height energy consumption by a cost function; and finally, pruning the initial path, removing redundant nodes, smoothing the trajectory by adopting a cubic spline interpolation method, ensuring curvature continuity, and generating an optimal path meeting the flight performance of the unmanned aerial vehicle. According to the invention, through an improved bidirectional minimum cost fast expansion random tree algorithm, dynamic association of obstacle constraint, target deviation and path cost in unmanned aerial vehicle path planning in a complex environment is represented, and a more efficient search convergence speed and better path quality are obtained. And rapid convergence and global optimization of the track of the unmanned aerial vehicle in a high-dynamic environment are realized.

The invention discloses a maritime unmanned aerial vehicle path planning method, which belongs to the technical field of path planning and comprises the steps of defining a three-dimensional grid map, preliminarily planning a path, optimizing the path and navigating the path. According to the scheme, bidirectional search is carried out through double search trees, the dynamic offset probability is calculated according to the obstacle density, the sampling point is generated, the nearest neighbor node is found, the gravitation of the target point, the repulsive force of the obstacle and the environmental force of the sea wind are combined to obtain the expansion resultant force, and then the adaptive expansion step length is obtained by combining the obstacle density. New nodes are generated and collision detection is carried out, and double search trees are connected to obtain an initial path, so that the efficiency, adaptability and safety of path planning are remarkably improved; according to the method, the Euclidean distance, the energy consumption value and the environment threat value are weighted and combined to obtain a comprehensive evaluation value of a path segment, an optimal jump node is selected for pruning, cubic spline interpolation smoothing processing is carried out to obtain a final path, and the flight requirement of the marine unmanned aerial vehicle can be better met.

The invention discloses a path planning method and system for a mobile robot, and relates to the technical field of path planning methods, and the method comprises the steps: determining a path planning model based on optimized point cloud data, environment complexity and the resolution of a visibility graph; in the determination of the path planning model, the optimized path is determined based on the two-way search of the starting point and the end point, the accuracy of the path planning model is ensured, rapid and accurate path planning is realized based on the two-way search of the starting point and the end point, and the accuracy of the optimized path is ensured. Collecting a plurality of path nodes, and dynamically optimizing each path node; according to the method, the obstacle features are collected according to detection of the optimized path, adaptive filtering is performed on the obstacle features, dynamic optimization of the path nodes and adaptive filtering of the obstacle features are compatible, so that the obstacles which have small influence on path planning can be adaptively filtered, and the calculation burden of path planning can be remarkably reduced.

The invention provides an improved point cloud registration method for a complex curved surface component, and relates to the technical field of computer vision and robots. The method comprises the steps of fixing the position of the complex curved surface component, performing measurement from two different visual angles by using a binocular structured light measurement system, and collecting surface information of the complex curved surface component; de-noising the source point cloud and the target point cloud according to local statistical characteristics of each point and neighboring points around the point; screening feature points, extracting feature point features of the source point cloud and the target point cloud, and determining feature corresponding point pairs; performing point cloud coarse registration according to the feature corresponding point pairs; acquiring local information of each point in the source point cloud and the target point cloud; determining a corresponding point set according to the local information and bidirectional search; and constructing a target function, and setting an iteration termination condition according to the corresponding point set to complete point cloud precise registration. According to the invention, precision registration of point clouds of different visual angles of a complex curved surface component can be realized, and the speed and precision of point cloud registration are improved.

The invention discloses a robot path planning method based on bidirectional beetle antennae search, and relates to the technical field of path search. The method comprises the following steps: S1, obtaining an obstacle, a feasible region, a starting point and an ending point, and forming an environment model; s2, placing a first long-horned beetle at the starting point, placing a second long-horned beetle at the terminal point, and enabling each long-horned beetle to always move towards the other long-horned beetle based on a beetle search algorithm to obtain a planned path; two-way search is carried out by placing a longhorn beetle at each of the starting point and the ending point, so that the speed of finding a feasible solution is increased, and the path planning efficiency is improved.

The invention provides a helicopter obstacle avoidance path planning method, which belongs to the technical field of aviation path planning, and comprises the following steps: marking obstacle points on an elevation grid map, and filling by adopting a water cup filling method; acquiring an end point meeting the constraint condition, and when the starting point or the end point is a filling barrier point, diffusing the filling barrier point cancelling from the starting point or the end point; a two-way A * algorithm is adopted for synchronous expansion, the obstacle avoidance path of the helicopter is searched, the obstacle avoidance path of the helicopter in a three-dimensional space is planned by designing the total cost of the starting point direction and the end point direction, an adjacent node list is generated, and new adjacent nodes are generated through a wall winding method; and when the two-way search encountering is carried out, backtracking is carried out along father nodes in two directions from the encountering node until the starting point and the ending point, and backtracking paths in the two directions are combined to obtain a complete path from the starting point to the ending point. According to the invention, the problem of low efficiency of a helicopter obstacle avoidance path search algorithm in a complex three-dimensional space is solved.

The embodiment of the invention discloses a moving body track planning method and device, electronic equipment and a storage medium, and relates to the technical field of track planning, and the method comprises the steps: carrying out the geometric boundary expansion and minimum distance detection of a moving body, and obtaining the environment complexity of a starting point and an end point of the moving body, the method comprises the steps of obtaining an initial trajectory and constructing kinematic constraints based on a mixed A * algorithm of a bidirectional search strategy, constructing a safety corridor based on the initial trajectory to obtain obstacle avoidance constraints, calculating environment complexity of each trajectory point according to the safety corridor, classifying the trajectory points to obtain sampling time constraints, and obtaining final constraints according to the constraints. Using the initial trajectory as an initial solution to obtain a planned trajectory, performing transverse and longitudinal decoupling on the planned trajectory to respectively obtain transverse and longitudinal state equations, and performing tracking control on the moving body by adopting a model predictive control (MPC) algorithm. According to the method, the problems of low calculation efficiency, infeasible trajectory planning result and poor optimization process convergence in the prior art are solved.

The invention discloses a global path planning method for a power converter valve inspection operation robot, and the method comprises the steps: obtaining the spatial point cloud data of a converter valve region through a three-dimensional structured light camera, and constructing a two-dimensional environment map containing obstacle expansion information through projection and rasterization; generating a snakelike or spiral full-coverage global theoretical path according to the scanning step length of the camera; in the inspection execution process, the feasibility of the path section is judged through discrete sampling and continuous collision detection; local obstacle avoidance adopts an improved A * algorithm, forward and backward bidirectional search is simultaneously executed in a limited local search area, and a dynamic heuristic function of a coverage potential item and a safety score item is introduced. And finally, splicing the local path and the theoretical path to form a global feasible inspection path. The method effectively solves the problem that the coverage rate and the obstacle avoidance efficiency are difficult to consider in a complex converter valve hall environment in a traditional method, and is particularly suitable for automatic inspection of equipment-intensive scenes such as an electric power converter valve hall.

The invention provides a robot safety interaction path planning method based on a dynamic danger criterion and bidirectional search, and the method comprises the steps: introducing the dynamic danger criterion in a planning stage, carrying out the quantification of a potential collision risk between a robot and a person, and comprehensively considering the inertia and rigidity of the robot, and the influence of the relative distance between the robot and the person; in combination with forward global search and reverse local search, a safe path is quickly generated in a complex environment. The forward search preferentially reduces danger indexes, the reverse search reversely avoids obstacles from a target position, the path continuity is ensured through an attitude potential energy function, and the local convergence problem of a traditional planning algorithm in an obstacle dense area is solved; the system also combines a real-time monitoring and safety control module, fuses a danger criterion, an obstacle potential field and a target potential field, and supports online path optimization in a dynamic environment so as to ensure that the robot can be always in a safe state in a final interaction process.

The invention discloses a path planning method for an underwater vehicle to pass through a narrow water channel, which can solve the problem of low path planning efficiency when the underwater vehicle operates through the narrow water channel. Firstly, an underwater vehicle operation space is selected, a marine environment feature model is established, and a final marine environment model is obtained; and defining the narrow water channel area through which the underwater vehicle passes by using the final marine environment model, and obtaining a grid chart of the task area. And based on the grid map of the task area, performing expansion processing and data simplification processing on the obstacles in the navigation obstructing area. And acquiring a global path by adopting eight-direction two-way A * search based on the processed grid map of the narrow water channel. And based on the grid map, carrying out real-time dynamic collision avoidance on the global path by adopting a DWA algorithm to carry out local path planning. Based on a DWA algorithm, an APF algorithm is adopted to carry out auxiliary real-time dynamic collision avoidance so as to carry out local path planning. And two actual sailing routes are obtained by adjusting the sailing direction and the safety coefficient in the DWA algorithm.

The invention relates to a medical glass bottle defect detection method and device, in particular to a medical glass bottle defect detection method and device based on machine vision, and aims to solve the problems that missing detection, false detection and the like are prone to occurring in an existing brown medical glass bottle detection technology. A medical glass bottle is conveyed to a rotary table through mechanical transmission equipment and rotates at a constant angle, a section contour image of the glass bottle is obtained, after the image is corrected through perspective transformation, background segmentation is achieved through an improved DeeplabV3 + network, an outer contour is extracted in combination with an improved Canny algorithm, and a three-dimensional point cloud is reconstructed. A KN-4PCS algorithm is used for coarse registration and an ICP algorithm is used for fine registration, contour difference points are calculated through bidirectional Kd-tree search, and size out-of-tolerance is judged. And meanwhile, the defects of the bottle body, the bottle opening and the bottle bottom are detected by utilizing an improved YOLOv7 algorithm. And the medical glass bottles with defects or unqualified sizes are removed through the removing mechanism. The overall defect detection of the medical glass bottle is completed.

The invention relates to the technical field of artificial intelligence, in particular to an operator fusion method, device and equipment and a computer readable storage medium, and the method comprises the steps: determining whether a to-be-split current node and an upstream node which are different in data arrangement type and are directly connected exist in a computational graph or not; if yes, performing downstream one-way search by taking the current node as a starting point to construct a first sub-graph containing a first associated node, and performing two-way search by taking the upstream node as a starting point to construct a second sub-graph containing a second associated node; taking the first sub-graph and the second sub-graph as a new calculation graph, and performing iterative detection until no direct connection node with different data arrangement exists; and if a plurality of first sub-graphs exist, combining the first sub-graphs with the same graph attributes to obtain a third sub-graph so as to realize operator fusion. The invention aims to improve the efficiency of operator fusion.

The invention relates to a three-dimensional track planning method based on a bidirectional APF-RRT *-CPO algorithm, and the method comprises the following steps: constructing a three-dimensional map, and creating two random trees T1 and T2; two-way search is adopted, sampling points are generated in the random trees T1 and T2 respectively, nodes closest to the sampling points are found respectively, and potential field force of the self-adaptive APF algorithm is utilized to guide generation of new nodes; rewiring is carried out, a nearest node connected with the rewiring is found, and nodes which do not collide are added into the random tree; connecting all nodes of the random tree to generate an initial track; and searching the shortest distance between the nodes by taking the track points of the initial track as the temporary starting point and terminal point, calculating a total cost function, and generating an optimal track according to the minimum cost position. The method is short in flight path searching time and high in convergence speed, and the flight path planning speed of the unmanned aerial vehicle is greatly increased in a complex three-dimensional environment.

The invention provides a chemical safety emergency dispatching command method and device, and belongs to the technical field of chemical safety emergencies, and the method comprises the steps: obtaining the operation state information of chemical production equipment and the real-time information of an emergency resource point; determining hidden danger factors of different position nodes in the chemical production network according to the operation state information, and further determining corresponding emergency resource types; modeling the chemical production network into a weighted graph; the hidden danger nodes are used as starting points, corresponding emergency resource types are used as constraints, layer-by-layer outward expansion is carried out according to network levels, bidirectional search is carried out from the hidden danger nodes and resource nodes, and edge weights are dynamically updated; determining response time according to the edge weight, optimizing the resource matching degree, the response time and the resource quantity as a multi-objective function, and generating an optimal path set; and performing scheduling command based on the optimal path set. Therefore, emergency resources can be ensured to arrive at an accident site at the fastest speed, and reasonable distribution and efficient utilization of the emergency resources are ensured.

The application belongs to the technical field of mechanical arm motion planning, and provides a kind of adaptive sampling and multi-channel expansion mechanical arm path planning method, first, the environment obstacle set is constructed, the point to obstacle distance, the clearance configuration and the segment level minimum clearance are established, and the unified safety criterion is constructed;Subsequently, according to the approximation state of bidirectional search tree, the sampling window is adaptively updated, and the random reference sample is generated in combination with the stagnation detection mechanism;On this basis, the hierarchical multi-channel expansion mechanism of direct pointing expansion, bypass expansion, potential field wall expansion and global bottom expansion is used to generate candidate new nodes;When a feasible new node is generated, it is connected to the current search tree and real meeting detection is carried out, if the double trees are successfully connected, path backtracking, splicing, simplification and full path safety review are further executed, and the final executable path is output.The application can improve the safety, search efficiency and environmental adaptability of the path planning of the mechanical arm in the complex limited environment.

The invention discloses a robot real-time online path planning method based on bidirectional search and node directions, and belongs to the technical field of path planning. The online path planning method solves the problems that an existing online path planning method is low in convergence speed and poor in planned path smoothness in a complex environment. According to the invention, by introducing a bidirectional search sampling method, the efficiency of path planning in a complex scene can be significantly improved, and when the robot moves on a planned path, whether a better path exists at each position at other time can be judged in real time according to a historical search result. By introducing a weighted node selection strategy, the problem of poor smoothness of a planned path can be improved, and the quality of path planning in a complex scene is remarkably improved. The method can be applied to real-time online path planning of the robot.

The application discloses a kind of global path planning methods for power converter valve inspection operation robot, obtains the spatial point cloud data of converter valve area using three-dimensional structured light camera, constructs two-dimensional environment map containing obstacle inflation information by projection and rasterization;According to the snake shape or spiral shape global theoretical path of full coverage generated by camera scanning step;During inspection execution, the feasibility of path segment is judged by discrete sampling and continuous collision detection;Local obstacle avoidance uses improved A* algorithm, simultaneously executes forward and backward bidirectional search in the limited local search area, and introduces the dynamic heuristic function of coverage potential item and safety score item.Finally, the local path and theoretical path are spliced, and the global feasible inspection path is formed.The application effectively solves the problem that the coverage rate and obstacle avoidance efficiency of traditional method are difficult to consider under complex converter valve hall environment, and is especially suitable for automatic inspection in equipment-intensive scenes such as power converter valve hall.

This application provides a warehouse picking method, device, electronic device, and computer-readable storage medium, relating to the field of warehouse picking. The method includes obtaining an order for goods to be picked; the order for goods to be picked includes at least one type of goods to be picked; performing a bidirectional search based on the warehouse location information of the goods to be picked to generate a picking plan path; the picking plan path is a shortest path that passes through all the goods to be picked in the order for goods to be picked; and picking the goods to be picked in the order for goods to be picked according to the picking plan path. This application can significantly shorten the picking path and improve picking efficiency.

The invention discloses an RT-RRT * global path planning method based on L2 greedy sampling, and belongs to the technical field of automatic driving vehicle path planning. According to the method, the problems of long time consumption and low convergence speed of path search in path search by the existing method are solved. According to the bidirectional search framework disclosed by the invention, two independent search trees are simultaneously constructed from a starting point and a target point to explore a state space, an L2 greedy sampling set is designed, and a small elliptical sampling region is defined around a state with the maximum heuristic estimation cost in a current solution path; according to the method, an L2 greedy sampling set is introduced to guide the search direction of two trees, so that the overall convergence speed of the algorithm can be increased, and the problems of long time consumption and low convergence speed of path search of an existing method are solved. And in combination with road environment constraints, obstacle constraints and a new node expansion strategy, the safety and smoothness of the planned path are enhanced. The method can be applied to path planning.

The invention belongs to the technical field of robot path planning, and particularly relates to a sampling optimal path planning method fusing adaptive heuristic and symmetric bidirectional search, which comprises the following steps: based on an improved EIT * algorithm, realizing symmetric bidirectional search by constructing a quadruple structure of a forward tree, a forward inertia tree, a reverse tree and a reverse inertia tree; in the random geometric graph approximate construction stage, an elliptical information set is adopted to limit a sampling area, and bidirectional edges are defined by connecting radius r or k neighbors; a middle encounter strategy is adopted, so that forward inert search and reverse inert search meet in the middle and share heuristic information, and the search efficiency is improved; the forward search and the reverse search respectively utilize heuristic information provided by inert search of the opposite side to carry out path exploration, the total cost is calculated after meeting and is continuously optimized, and finally an optimal path is obtained through pruning; according to the method, the initial solution obtaining speed is effectively increased, the path quality is improved, and a higher planning success rate is shown in a complex environment.

The invention relates to a quick single-source path-finding algorithm based on bidirectional search, and belongs to the technical field of robot path planning. Aiming at the problems of path congestion and deadlock and poor dynamic adaptability in AGV path planning in warehouse logistics, the invention provides a bidirectional collaborative search mechanism: reversely executing a Dijkstra algorithm from a target point to pre-calculate the cost of the shortest distance from each node to the target, and selecting a node dynamic expansion path with the minimum comprehensive distance and steering cost in real time in combination with forward search; the path to the target point can be quickly generated by quickly combining the information obtained by the reverse search with the state of the AGV during the forward search. According to the algorithm, the next node is generated in real time during AGV movement, the turning frequency is reduced, and the real-time performance and efficiency of path planning in a complex environment are improved.