159 results about "Mobile robot navigation" patented technology

A volumetric sensor for mobile robot navigation to avoid obstacles in the robot's path includes a laser volumetric sensor on a platform with a laser and detector directed to a tiltable mirror in a cylinder that is rotatable through 360° by a motor, a rotatable cam in the cylinder tilts the mirror to provide a laser scan and distance measurements of obstacles near the robot. A stereo camera is held by the platform, that camera being rotatable by a motor to provide distance measurements to more remote objects.

The invention, which belongs to the technical field of mobile robot navigation, provides a mobile robot obstacle avoidance method based on a DoubleDQN network and deep reinforcement learning so that problems of long response delay, long needed training time, and low success rate of obstacle avoidance based on the existing deep reinforcement learning obstacle avoidance method can be solved. Specialdecision action space and a reward function are designed; mobile robot trajectory data collection and Double DQN network training are performed in parallel at two threads, so that the training efficiency is improved effectively and a problem of long training time needed by the existing deep reinforcement learning obstacle avoidance method is solved. According to the invention, unbiased estimationof an action value is carried out by using the Double DQN network, so that a problem of falling into local optimum is solved and problems of low success rate and high response delay of the existing deep reinforcement learning obstacle avoidance method are solved. Compared with the prior art, the mobile robot obstacle avoidance method has the following advantages: the network training time is shortened to be below 20% of the time in the prior art; and the 100% of obstacle avoidance success rate is kept. The mobile robot obstacle avoidance method can be applied to the technical field of mobilerobot navigation.

The invention discloses a mobile robot navigation method based on a binocular camera and a two-dimensional laser radar. The mobile robot navigation method includes the steps that a two-dimensional grid map model is built; a mobile-robot pose model is built; a laser radar data model is built; a joint calibration model of the binocular camera and the laser radar is established; route planning and obstacle avoidance navigation are carried out with the Dijkstra algorithm and the A-Star algorithm. According to the mobile robot navigation method, the relative position and the relative orientation ofthe binocular camera and the two-dimensional laser radar can be determined with the triangle joint calibration method; in the known-environment two-dimensional grid map environment, the binocular camera can be used for detecting barriers higher than and lower than the two-dimensional laser radar plane, the barriers can be mapped into an environment map, through the shortest path algorithm and heuristic search optimization, a navigation path line capable of avoiding the barriers is obtained, and a mobile robot can reach the destination along a path capable of avoiding the barriers in the movement process.

The invention discloses a collision avoidance planning method for mobile robots based on deep reinforcement learning in a dynamic environment, and belongs to the technical field of mobile robot navigation. The method of the invention includes the following steps of: collecting raw data through a laser rangefinder, processing the raw data as input of a neural network, and building an LSTM neural network; through an A3C algorithm, outputting corresponding parameters by the neural network, and processing the corresponding parameters to obtain the action of each step of the robot. The scheme of the invention does not need to model the environment, is more suitable for an unknown obstacle environment, adopts an actor-critic framework and a temporal difference algorithm, is more suitable for a continuous motion space while realizing low variance, and realizes the effect of learning while training. The scheme of the invention designs the continuous motion space with a heading angle limitationand uses 4 threads for parallel learning and training, so that compared with general deep reinforcement learning methods, the learning and training time is greatly improved, the sample correlation isreduced, the high utilization of exploration spaces and the diversity of exploration strategies are guaranteed, and thus the algorithm convergence, stability and the success rate of obstacle avoidance can be improved.

The invention discloses a virtual scanning and ranging matching-based AGV laser SLAM method, and relates to mobile robot navigation and positioning. The method comprises a grid map representation and establishment method, a virtual scanning and matching positioning method and an algorithm instantaneity improving method. The method using a contour traversal matching principle comprises the specific steps of scanning a map by adopting a virtual laser radar at each traversal position, directly comparing virtually scanned data with data of a current laser radar, finding out optimal position information of an AGV robot and then incrementally building the map. For the problems that most of existing laser SLAM algorithms are aimed at a low-precision sensor, and the filtering, estimating and optimizing stability and the positioning accuracy cannot be absolutely ensured and the application requirements of the industrial AGV robot cannot be easily met, the problem of preliminary construction and calibration in navigation by adopting a reflector board and a triangulation principle can be solved by adopting multi-GPU parallel processing and changing an initial propulsion position of virtual ranging, and the flexibility, the reliability and the precision are improved.

The embodiment of the invention relates to a method and a device applied to indoor location of a mobile robot. The method comprises the following steps: estimating the current location of the mobile robot according to the indoor location technique of the mobile robot; accessing a reference object database and selecting a real location and relevant reference images from the database; navigating the mobile robot to the selected real location; shooting images by using a camera matched with the mobile robot to obtain shot images; comparing reference object images included in the shot images with reference object images included in the selected reference images; and determining location errors generated by the indoor location technique of the mobile robot according to comparison results. Through the method and the device applied to indoor location of the mobile robot according to the embodiment of the invention, the indoor location of the mobile robot can be improved.

The invention discloses a mobile robot visual positioning and navigating method based on an ArUco code and belongs to the fields of computer visual positioning and mobile robot navigation. The mobilerobot visual positioning and navigating method comprises the following steps: map creation; route creation; and positioning and navigating. By using the method disclosed by the invention, not only arethe defects such as strong dependency on auxiliary physical condition, high environmental requirement and inflexibility of a traditional visual navigation way overcome, but also a traditional visualnavigation method only suitable for a ground mobile robot is widened to a three-dimensional space, so that the mobile robot visual positioning and navigating method is very suitable for positioning and navigating an aerial unmanned aerial vehicle; and the mobile robot visual positioning and navigating method has various advantages such as high precision, high flexibility and wide application rangeand can be used for autonomously positioning and navigating intelligent mobile equipment such as mobile robots and unmanned aerial vehicles in an indoor environment, so that the application scenes are widened.

The invention discloses a charging method of a mobile robot. The charging method comprises the following steps of S1, detecting a charging seat by a laser radar or an infrared signal when the mobile robot is in a non charging-back state; S2, adding position information of the mobile robot when the charging seat is detected or position information of the charging seat into a map by the mobile robot; S3, seeking a position of the charging seat by the mobile robot according to the map; and S4, achieving aligned charging-back with the charging seat by the mobile robot according to an infrared charging-back signal. The position information of the mobile robot or the position information of the charging seat when the charging signal is detected is added into the map, the charging seat is sought according to the map information when charging-back is required by the mobile robot, the mobile robot can be rapidly navigated to a part near to the charging seat, alignment is performed by an infrared positioning mode, and the charging method has the advantages of rapid charging-back and high alignment accuracy. Moreover, the invention also discloses a charging system of the mobile robot.

The invention discloses a multi-grid value navigation method based on the robot pose and an application thereof and belongs to the technical field of robot navigation. The method is characterized in that the accessible area of the robot in the grid map is expanded, and connectivity of topology maps is enhanced, moreover, through the indoor wall construction autonomous mobile robot navigation method based on the BIM information, working points are calculated based on the extracted BIM information, according to the position information of the working points and the order of work tasks, the multi-grid value navigation method based on the robot pose is utilized to set local navigation routes between the adjacent working points; real-time positioning is performed, and the obstacle information around an indoor wall construction autonomous mobile robot is detected, the grid map is updated, and the local navigation routes between the adjacent working points are reset based on the updated gridmap. The method is advantaged in that navigation work efficiency and accuracy are improved.

The invention discloses a mobile robot navigation control method based on laser data, comprising the steps of 1, adding and setting a path in a global map; 2, acquiring real-time positioning information of a mobile robot through iterative matching algorithm; 3, planning a shortest path required by navigation of the mobile robot through shortest path algorithm; 4, navigating the mobile robot to run along the planned path through left and right wheel differential speed control; 5, enabling the mobile robot to park accurately at a target parking point on the path through deceleration control. The mobile robot navigation control method based on laser data has the advantages that data acquisition is simple and needs no extra laying of a landmark, the cost is low, positioning precision is high, and the mobile robot is allowed to be autonomously positioned and navigate-controlled in indoor and outdoor environments.

A double-threshold mechanism is used for implementing the follow-me function of a mobile robot. Initially, a user comes to a mobile robot and turns on its follow-me function. Then, the mobile robot is in the follow-me mode or can simply be described as following the user. When the user moves away from the robot, the robot determines whether the distance between itself and the user exceeds a first distance threshold. If so, the robot starts moving to follow the user. Otherwise, the robot stays put. While following the user's movement, the robot continues to monitor the distance between itself and the user. When the robot determines that the distance between them is less than a second distance threshold—because the user has slowed down or stopped, for example—the robot stops moving. The second distance threshold is lower than the first distance threshold.

The invention relates to a construction method for a mixed map for the navigation of mobile robots. A whole area is divided into a plurality of local areas, and each local area has an overlapping area with one or more adjacent local areas thereof, visual marks are added in each overlapping area, the mobile robots respectively move in each local area and acquire data of all points in each local area; an occupied grid map principle is employed to have acquired data in each local area generate corresponding local measuring maps; a note is added on each local measuring map, the note comprises a switch node, and the switch nodes correspond to the visual marks of the overlapping areas; and same switch nodes of different local measuring maps are connected, and the topology structure of whole area is formed. According to the method, the reliability of the mixed map can be realized, management and maintenance of the maps can be facilitated, the adaptability and expansibility are realized, and reliable navigation information can be provided for the mobile robots.

The invention provides a mobile robot navigation method based on one point RANSAC and FAST algorithms. The method mainly includes that a robot adopts the FAST corner extraction algorithm for collecting surround environment information through a self-equipped camera; the one point RANSAC algorithm combined with extension Kalman filtering is adopted during a matching process for error matching remove; and obtained matching points are used for three-dimensional environment reconstruction and map creation. Aiming at a problem that the number of iterations is too large in the prior art employing the RANSAC algorithm, prior information obtained in a filtering phase is utilized fully. In this way, the number of iterations of the algorithm is reduced effectively and the algorithm robustness is guaranteed at the same time, so that feature matching is implemented. Problems that a traditional mobile robot is poor in monocular vision navigation real time performance and robustness and the number of iterations of the RANSAC algorithm is too large are solved.

The invention discloses a robot path planning method based on an ANFIS fuzzy neural network, and mainly solves the problems of complex trap path reciprocating and path redundancy in conventional reactive navigation. The method comprises the following steps: to begin with, establishing a kinematic model for a mobile robot; providing a mobile robot navigation controller based on the fuzzy neural network by means of autonomous learning function of the neural network and fuzzy reasoning ability of the fuzzy theory; constructing a Takagi-Sugeno fuzzy inference system based on an adaptive fuzzy neural network structure and serving the Takagi-Sugeno fuzzy inference system as a reference model for local reaction control of the robot; the fuzzy neural network controller outputting offset angle and operation speed in real time, and adjusting the migration direction of the mobile robot online to enable the mobile robot to be able to adjust speed automatically and approach the goal collisionless; and through an improved virtual target method, selecting an optimal path capable of allowing the robot to escape a trapping state.

The invention discloses a navigation deviation correction method for a mobile robot, comprising: obtaining a navigation map of the working environment of the mobile robot and a planned movement route; collecting environmental information of the mobile robot during the moving process, and obtaining current pose information of the mobile robot according to the environmental information; According to the mobile planning route and the current position and orientation information of the mobile robot, the navigation deviation correction control information is obtained; according to the navigation deviation correction control information, the deviation generated when the mobile robot moves is corrected in real time. In addition, the invention also discloses a navigation deviation correction device for a mobile robot. Through the present invention, the deviation generated during the movement of the mobile robot can be corrected in real time, so that the mobile robot can accurately arrive at the destination. The navigation deviation correction method and the navigation deviation correction device of the present invention are simple and practical, low in implementation cost and high in accuracy.

Disclosed herein are an apparatus and method for generating a topological map for navigation of a robot. The method for generating a topological map for navigation of a robot, performed by the apparatus for building the topological map for the navigation of the robot, includes calculating the physical size of a single pixel on a metric map of a space in which a mobile robot is to navigate, extracting the physical coordinates of the pixel on the metric map, building node data and edge data for the navigation of the mobile robot using the physical coordinates, and generating a topological map for the navigation of the mobile robot based on the built node data and the built edge data.

Methods and systems for navigating a mobile robot through a crowded pedestrian environment by selecting and following a particular pedestrian are described herein. In one aspect, a navigation model directs a mobile robot to follow a pedestrian based on the position and velocity of nearby pedestrians and the current and desired positions of the mobile robot in the service environment. The mobile robot advances toward its desired destination by following the selected pedestrian. By repeatedly sampling the positions and velocities of nearby pedestrians and the current location, the navigation model directs the mobile robot toward the endpoint location. In some examples, the mobile robot selects and follows a sequence of different pedestrians to navigate to the desired endpoint location. In a further aspect, the navigation model determines whether following a particular pedestrian will lead to a collision with another pedestrian. If so, the navigation model selects another pedestrian to follow.

The invention discloses a navigation and positioning performance testing device and method for an autonomous mobile robot. The device comprises a testing base station, a mobile station and an off-line data analysis and processing system. The testing base station is constituted by a differential GPS base station, a data transmission transmitter and a power management module. The mobile station is composed of a GPS receiver module, a data transmission receiver, an on-line data acquisition and processing system and a power management module. The testing base station and the mobile station constitute a high-precision movement track testing hardware device, and motion path information data of the robot are recorded in real time. According to the method, the tested robot performs strict linear motion, self-spinning motion around the center and Z-shaped motion and follows a testing track constructed according to the international standard ISO-3888-1. Statistics and analysis are conducted on position information of the robot through the off-line data analysis and processing system, and robot navigation and positioning performance assessment is achieved. The navigation and positioning performance testing device and method have the advantages of being high in testing precision, good in dynamic performance and the like.

The invention discloses a mobile robot navigation method. The method comprises the following steps that 100, a raspberry pi control panel is installed on a chassis of a mobile robot; 200, a bottom layer driver file of an ROS system is transplanted; 300, a special mobile robot driver file is transplanted; 400, source codes of a navigation algorithm are complied. A robot navigation algorithm can be conveniently modified/updated.

The invention relates to a wheel type mobile robot navigation method based on an IHDR self-learning frame. The method comprises following steps: firstly, a robot is guided to move, and in the process, the robot is controlled to acquire current image information and speed values in real time via a kinect sensor and a handle; then the IHDR algorithm is employed to establish a mapping relation between image input and speed value output and store a lot of acquired data in the robot in the manner of a "knowledge" tree-shaped structure; finally, the robot can search in a knowledge base according to the current image information without human intervention so that self navigation is realized. According to the method, the robot has the same thinking mode with people, a cognitive ability is reflected, the navigation is not realized via a mathematical model in a conventional manner, and the intelligent degree is higher.

The invention aims to provide a navigation and positioning method for an indoor mobile robot. Two positioning encoding disks of which encoder axes are not parallel and which are distributed at any angle and an uniaxial optical fiber gyroscope are used for positioning, the positioning encoding disks are installed on the indoor mobile robot, positioning data are directly transmitted to a control system in time, and the data which are fed back by encoders are processed by the control system through a positioning algorithm, so that coordinates of the robot can be positioned. According to the navigation and positioning method, on the basis of the positioning encoding disks and the uniaxial optical fiber gyroscope, i.e. two optical encoders provided with omnidirectional wheels are vertically distributed, the indoor mobile robot belongs to a mechanical device, compared with a wireless sensing response network system, a GPS (Global Position System) navigation and positioning system and the like based on electronic components, a measuring device provided by the invention is simple, the implementation is convenient, and the mechanical device has high stability; the positioning accuracy of the robot can be controlled within 0.5% through experimental debugging for about five times, and high positioning accuracy is obtained.

The invention discloses a binocular vision-based method for detecting a barrier in an indoor shadow environment. A binocular vision system is adopted. The method comprises the steps of (1) removing a shadow, extracting color saturation of an original image, performing fusion with image information subjected to shadow removal, and adaptively adjusting and increasing grayscale difference of the ground and the barrier through environmental brightness information; (2) filling a ground region of a fused image by utilizing a seed filling algorithm, and obtaining a barrier region through threshold segmentation, corrosion and expansion operations; and (3) performing matching in a right camera image by taking the obtained barrier region as a template through applying the binocular vision system to calculate parallax of a central point, and calculating three-dimensional coordinates of the central point and the width and distance of the barrier. The invention furthermore discloses a binocular vision-based apparatus for detecting the barrier in the indoor shadow environment. According to the method and the apparatus, the barrier can be completely extracted in the indoor shadow environment; and the method and the apparatus are simple and efficient, have relatively good real-time property and relatively high precision, and are suitable for mobile robot navigation and barrier avoidance.