33 results about "Dynamic window approach" patented technology

The invention discloses an intelligent path planning method for a mobile robot, and the method comprises the steps: building a static two-dimensional grid map, and carrying out the global path planning through an improved ant colony algorithm; enabling the mobile robot sensor module to detect unknown obstacle information, calculating an obstacle motion trail and a robot motion trail, adopts an optimized dynamic window method to carry out local dynamic obstacle avoidance, and taking the current position of the robot as a starting point and a closest key node on a global planning path as a temporary target point to carry out dynamic obstacle avoidance; and enabling the robot to travel along the planned path and safely reach the destination. According to the method, the actual problems of static obstacles and dynamic obstacles in the map environment are comprehensively considered, the heuristic function of the ant colony algorithm is improved, pheromone updating rules are adjusted for global path planning, the optimized dynamic window method is adopted for obstacle avoidance when the robot encounters the dynamic obstacles in the running process, and local path planning is completed; and the robot has higher practicability and research value in actual map operation.

Provided is an obstacle avoiding method based on a dynamic window and virtual target points. A robot is guided by virtual target points to advance, and a motion instruction of the robot is issued through a dynamic window, and thus, the robot can avoid an obstacle to reach a target point. First, a robot predicts the motion track of an obstacle according to the angle and distance information of theobstacle fed back by a sensor on the robot; then, multiple virtual target points are generated according to the track prediction of the obstacle, the motion state of the robot and the position of a real target point, and an optimal virtual target is screened out by considering the direction of the robot and the distance between the virtual target points and the real target point through an evaluation function; and finally, the robot generates a control instruction set of the robot for the next moment through a dynamic window approach according to the track prediction of the obstacle and the positions of the virtual target points, and screens out an optimal control instruction of the robot for the next moment by considering the direction and motion speed of the robot and the distance between the robot and the target point through an evaluation function.

The invention discloses an improved ant colony algorithm and dynamic window method-based hybrid algorithm applied to intelligent vehicle path planning. The algorithm can realize real-time obstacle avoidance, improve the path planning efficiency and realize automatic feedback control of an intelligent vehicle. According to the algorithm, after the Matlab simulation platform and the ROS-based intelligent vehicle platform are used for specifying a starting point and a target point on the premise that the obstacle is known, global path planning of the intelligent vehicle is completed, meanwhile,unknown obstacles are set on the grid map, local moving target points are planned, the local target points are tracked in real time, and local real-time obstacle avoidance of the intelligent vehicle is completed. According to the intelligent vehicle path planning method based on the improved ant colony algorithm and the dynamic window method, the turning angle of the path is smoothened, the path planning efficiency is improved, real-time obstacle avoidance of the intelligent vehicle is achieved, and good robustness and accuracy are achieved.

The invention discloses a robot and a navigation method thereof and a computer readable storage medium. The method includes the steps that target position information of the robot and two-dimensional point distribution of the current direction of the robot are obtained; the multi-dimensional feature value s of the robot is determined according to the obtained two-dimensional point distribution of the current direction of the robot; the control signal a of the robot is calculated through a model a equal to f (s, theta) according to the determined multi-dimensional feature value s of the robot and the theta value obtained through a machine learning method; navigation of the robot is achieved according to the calculated control signal a of the robot. According to the robot and the navigation method thereof and the computer readable storage medium, the control signal a of the robot is calculated through the model a equal to f (s, theta) through the theta value obtained through the machine learning method, and then navigation of the robot is achieved; relative to DWA (Dynamic Window Approach), TEB (Timed Elastic Band) and other local navigation algorithms in the prior art, the local navigation problem of the robot is effectively solved, the method is smooth in running, and efficiency is improved compared with an existing algorithm.

The invention belongs to the technical field of artificial intelligence and robot navigation, and particularly relates to a hybrid path planning method for an indoor mobile robot. According to the method, using a grid method for modeling the environment; searching a global path by using a bat algorithm as a global path planning method; optimizing the obtained global path, namely deleting redundantpath nodes, and reducing the length of the path and turning points of the path; designing an evaluation function of a dynamic window method, and bringing the global path into the evaluation functionof the dynamic window method to realize the combination of local planning and global planning; introducing an adaptive idea to dynamically adjust the coefficient of the evaluation function, and improving the dynamic obstacle avoidance capability of the fusion algorithm. The method can be applied to path planning of the indoor mobile robot, path planning and robot control are combined, and dynamicobstacle avoidance is considered under the condition that the global optimal path is guaranteed.

The invention discloses a vehicle obstacle avoidance method based on binocular vision and deep learning. The method comprises the following steps: obtaining an RGB image in front of a vehicle; acquiring a depth information map in front of the vehicle; predicting a drivable area segmentation result; carrying out information fusion to optimize a drivable area; removing the depth information of a drivable area part, and generating a depth obstacle front view; acquiring an obstacle distribution condition in a three-dimensional space in front of the vehicle, and obtaining an aerial view obstacle scatter diagram according to the obstacle distribution condition; performing density clustering on the aerial view obstacle scatter diagram, and removing noise; performing euclidean distance transformation on the aerial view obstacle scatter diagram, setting an adaptive threshold value, and dividing a front map into a safe driving area and a dangerous area; constructing a map for path planning through the aerial view safe driving area map and the field angle boundary information; performing obstacle avoidance path planning by using a dynamic window method in combination with a map for path planning; and calculating an expected speed and an expected angle according to the obstacle avoidance path track and issuing the same to a control system. The invention further provides the corresponding electronic equipment.

A track evaluation method provided by the invention comprises the following steps: acquiring N pieces of obstacle distance information; determining a target obstacle distance according to the N piecesof obstacle distance information; obtaining the distance between a terminal point of a preset path and the current position of a mobile robot, and taking the distance as a target point distance; forany to-be-evaluated track in a to-be-evaluated track set, determining a first weight corresponding to a first estimated distance and a second weight corresponding to a second estimated distance according to the target point distance and the target obstacle distance; according to the N pieces of obstacle distance information, determining a third weight corresponding to a target metric value, wherein the target metric value is used for indicating the distance between the to-be-evaluated track and an obstacle; and according to the first weight, the second weight and the third weight, obtaining anevaluation value corresponding to the to-be-evaluated track through a dynamic window method. Through the method, the evaluation accuracy of the to-be-evaluated track of the mobile robot in path planning can be improved.

The invention relates to an unmanned vehicle real-time global path planning method based on a dynamic window of a safety A* guide point. The method comprises steps that a node safety extension strategy and a path node quadratic optimization method are firstly provided on the basis of an A* algorithm, which are called as a safety A* algorithm, and an optimal safety virtual target point is rapidly found out through employing the safety A* algorithm; secondly, the virtual target point serves as a local target of a dynamic window method, speed sampling is conducted, path planning and obstacle avoidance are achieved, simulation is conducted in the MATLAB environment, the result shows that the unmanned vehicle can safely and stably avoid the obstacle under the guidance of the global path, and safety and stability of the method are verified.

The invention discloses a curvature consistency path planning algorithm based on an adaptive dynamic window method in a narrow channel environment, and belongs to the technical field of robot navigation. The method comprises the following steps: calculating the distance with a nearest obstacle, and if the distance D is smaller than a threshold value Dt, calculating a dynamic weight gamma of a speed item vel in a target function; calculating a curvature k of a track corresponding to each sampling speed, thereby calculating a curvature similarity factor Csim; and calculating values of a headingitem, a dis item and a vel item in the target function of a standard dynamic window method, and substituting the curvature similarity factor Csim and the dynamic weight gamma of the vel item to obtainan optimal motion trail and a corresponding execution speed at the next moment. According to the method, the problems that when the robot carries out local path planning in the narrow channel, the robot is prone to falling into local optimum, consequently, the terminal point cannot be reached, and the smoothness of the motion planning trajectory is poor are solved, and the reasonability and consistency of trajectory planning in the narrow channel environment are improved.

The invention relates to the field of autonomous navigation of unmanned ships, in particular to an unmanned ship layered obstacle avoidance method based on improved fruit fly optimization and a dynamic window method. The method comprises the following steps: S1, establishing an unmanned ship marine navigation geographical environment model based on an electronic chart; S2, completing global optimal path planning by adopting an improved fruit fly optimization algorithm; S3, establishing an environment model of the dynamic obstacle; and S4, using an improved dynamic window method to avoid the moving obstacle ship. According to the obstacle avoidance method for switching the dynamic obstacle avoidance mode and the static obstacle avoidance mode based on the judgment conditions, obstacles canbe effectively avoided during sailing, the algorithm is prevented from falling into a local optimal solution, the fuzzy control method is adopted for dynamically controlling the weight parameters of the optimal trajectory, and the trajectory prediction precision and efficiency are improved.

The invention discloses a dynamic path planning method for an improved particle swarm algorithm, which is characterized by comprising the following steps of: 1, generating a global path by adopting the improved particle swarm algorithm, and dividing the global path into n sections of local paths consisting of {S1G1, S2G2,... SnGn}; step 2, performing path planning on the local path S1G1 by using a dynamic window method DWA, including building a robot kinematics model, building a robot speed model, building an objective function of the dynamic window method DWA, selecting an optimal track according to the objective function, and performing path planning on the local path S1G1 by using the optimal track; step 3, performing local path planning on the local path S2G2 by using a dynamic window method DWA, and inheriting the course angle of the previous motion planning; 4, the dynamic window method DWA is repeatedly used for conducting local path planning on the paths S3G3,..., SnGn in sequence; and step 5, outputting a complete final path. The method has the characteristics of shortening the length of the planned path, improving the smoothness and real-time performance of the planned path and the like.

The invention provides a local path planning algorithm and device, which are used for path planning of mobile equipment that is provided with a sensor. The algorithm comprises the following steps: S1,judging whether the mobile equipment reaches a destination or not, and if so, ending a process, or if not, executing step S2; S2, acquiring a candidate gap set according to sensor data; S3, selectinga nearest gap from the candidate gap set as a target gap; S4, determining the moving direction of the mobile equipment according to the target gap; S5, controlling the mobile equipment to move alongthe target direction by adopting an improved dynamic window method; and S6, repeating the steps S1 to S5 until the mobile equipment reaches the destination or no feasible gap exists. The path of the mobile equipment is planned based on gaps and an improved dynamic window method so as to reduce the condition of local extreme value and improve obstacle avoidance performance.

The invention relates to a local path planning method based on a dynamic window method and suitable for an Ackerman model robot. The method comprises the steps of: firstly, acquiring the posture of the robot and the position of an obstacle according to a sensor carried by the robot, then calculating the coordinates of an obstacle avoidance starting point and coordinates and direction of a local target point, and determining a final speed space; meanwhile, segmenting the obstacle avoidance process of the robot to determine evaluation functions, judging the number of segments in segmented obstacle avoidance at present, and selecting the corresponding evaluation function for processing to obtain the speed corresponding to the minimum evaluation function value to serve as the optimal speed ofrobot operation, and repeatedly executing the obtained optimal speed, judging whether the robot reaches a local target point or not according to the current position of the robot, and if so, ending obstacle avoidance; conducting local path planning according to the size and position of the obstacle, the robot speed and the minimum turning radius, and promoting the robot to move in an optimal pathon the premise that the robot can avoid the obstacle and can return to a global path in a smooth direction after obstacle avoidance is finished.

The invention provides an unmanned surface vehicle path planning method which is an unmanned surface vehicle path planning method based on a dynamic window method and an optimized LPA * algorithm in an environment with moving obstacles. The method comprises the following steps: obtaining a global remote sensing image of a water surface environment through aerial photography of an aircraft, carrying out classification and binarization processing on the aerial photography image, converting the image into a grid map, carrying out global path planning by utilizing an LPA * algorithm of an optimized distance heuristic function based on the global grid map of the environment to obtain an initial path, and carrying out interpolation smoothing processing on the path; and based on a dynamic window method, constructing an evaluation function with a globally optimal path, and performing local path dynamic planning to achieve a local dynamic planning capability on the premise of the globally optimal path. Simulation results show that the method can obtain a smooth path better than a traditional heuristic A * algorithm, and can also realize good obstacle avoidance performance in a moving obstacle environment; moreover, the method is integrated with the structure of an LPA * global programming algorithm, and overcomes the defect that a dynamic window method is liable to fall into local optimum.

The invention relates to a mobile robot motion planning method and system based on Mecanum wheels. The method comprises the following steps of: creating a grid map based on a working space; based on the grid map, adopting an A * algorithm to carry out global path planning on the mobile robot based on Mecanum wheels to obtain a global path, wherein the global path is a path of the mobile robot from a starting point to an end point, and the global path comprises a plurality of nodes; by taking each node on the global path as a target point, performing local trajectory planning on the mobile robot based on an improved dynamic window method, and tracking the global path. The method is high in global path search efficiency, good in local obstacle avoidance effect and high in robustness, and can help the Mecanum wheel-based mobile robot to autonomously plan paths, flexibly avoid obstacles and safely and stably move based on the field environment, so that the flexibility of the movement is improved, and the autonomy of the movement is enhanced.

The invention discloses a mobile robot obstacle avoidance method based on pedestrian prediction, and the method refers to a thought that people can analyze the movement trend of moving obstacles when facing the moving obstacles, and then carry out the avoidance in advance. The moving robot is divided into a long and narrow corridor section and a wide hall section according to different scenes in the running process of the moving robot, when the robot encounters pedestrians in front, a social force model is used for predicting the walking track of the robot in the next period of time, an inactive avoiding mode is adopted in the long and narrow corridor section, and the speed of the robot is adjusted according to the predicted positions of the pedestrians; when the pedestrian is located in a hall section, an active avoidance strategy is adopted, a dynamic window method (DWA) is improved, and an evaluation item for predicting the direction of the pedestrian is newly added in an original evaluation function, so that the improved algorithm can avoid the position to which the pedestrian wants to walk in advance. According to the mobile robot obstacle avoidance method provided by the invention, the walking intention of the pedestrian is added to the obstacle avoidance decision of the robot, and the dynamic obstacle avoidance efficiency of the mobile robot is improved.