650 results about "Robotic navigation" patented technology

A method and apparatus of improving the navigation performance of robot are provided. The navigation method using a virtual sensor includes: generating information on positions of obstacles, the information which is estimated to be generated by virtual sensors that are virtually present, based on information on positions of the obstacles, which is generated by physical sensors; and controlling a movement of a robot according to the information on the positions of the obstacles. It is possible to overcome limits in the number and arrangement of previously installed physical sensors.

The invention provides a robot navigation method and device based on multi-sensor data fusion. The method comprises the following steps: establishing a map of a total environment according to data acquired by a laser radar sensor and data of an encoder; acquiring the current position of a robot in the map of the total environment in real time according to data acquired by a laser radar sensor, an accelerometer sensor, a gyroscope sensor and a magnetometer sensor, the map of the total environment and the data of the encoder; acquiring a planned route of the robot from the current position to a targeted position in real time according to the map of the total environment and the current position of the robot; and controlling the robot to keep away from barriers during movement by the data which are acquired by the laser radar sensor, a deep camera, an ultrasonic sensor and an infrared sensor and the data of the encoder. On the basis of reasonable utilization of the sensors to realize navigation of the robot, the method is flexibly applied to various scenes, costs are considered, and the autonomous navigation effect can be achieved.

The invention is related to methods and apparatus that use a visual sensor and dead reckoning sensors to process Simultaneous Localization and Mapping (SLAM). These techniques can be used in robot navigation. Advantageously, such visual techniques can be used to autonomously generate and update a map. Unlike with laser rangefinders, the visual techniques are economically practical in a wide range of applications and can be used in relatively dynamic environments such as environments in which people move. One embodiment further advantageously uses multiple particles to maintain multiple hypotheses with respect to localization and mapping. Further advantageously, one embodiment maintains the particles in a relatively computationally-efficient manner, thereby permitting the SLAM processes to be performed in software using relatively inexpensive microprocessor-based computer systems.

Methods and systems are disclosed for autonomously building a predictive model of outcomes. A most-predictive set of signals S_k is identified out of a set of signals s₁, s₂, . . . , s_D for each of one or more outcomes o_k. A set of probabilistic predictive models ô_k=M_k (S_k) is autonomously learned, where ô_k is a prediction of outcome o_k derived from the model M_k that uses as inputs values obtained from the set of signals S_k. The step of autonomously learning is repeated incrementally from data that contains examples of values of signals s₁, s₂, . . . , s_D and corresponding outcomes o₁, o₂, . . . , o_K. Various embodiments are also disclosed that apply predictive models to various physiological events and to autonomous robotic navigation.

The invention provides a robot adaptive grabbing method based on deep reinforcement learning. The method comprises the following steps: when distanced a certain distance away from an object to be grabbed, a robot obtaining a picture of a target through a pick-up head in the front, then according to the picture, calculating position information of the target by use of a binocular distance measurement method, and applying the calculated position information to robot navigation; when the target goes into the grabbing scope of a manipulator, taking a picture of the target through the pick-up head in the front again, and by use of a DDPG-based deep reinforcement learning network trained in advance, performing data dimension reduction feature extraction on the picture; and according to a feature extraction result, obtaining a control strategy of the robot, and the robot controlling a movement path and the posture of the manipulator by use of a control strategy so as to realize adaptive grabbing of the target. The grabbing method can realize adaptive grabbing of objects which are in different sizes and shapes and are not fixedly positioned and has quite good market application prospect.

The invention discloses an indoor movable robot real-time navigation method based on visual information correction, comprising the following steps of: (1) initializing a robot navigation system, and starting a mile meter and a visual sensor; (2) obtaining the current position of a robot by the mile meter and the visual sensor and forming a kalman filter so as to obtain a mile meter filtration estimation value and a visual sensor filtration estimation value; (3) fusing the mile meter filtration estimation value and the visual sensor filtration estimation value; and (4) resetting parameters. The invention sufficiently utilizes the respective advantages of visual information and mile meter information and combines the precision of the visual information and the real-time property of the mile meter information; the invention utilizes the mile meter self information to carry out the recurrence computation to obtain navigation data at most of the time, thereby ensuring the real-time requirement of the navigation system; in addition, the invention also utilizes the visual information to correct the accumulated errors generated in the dead reckoning of the mile meter, thereby greatly enhancing the accuracy of the navigation system.

The invention relates to a service robot autonomous navigation method based on a deformable topological map in the technical field of robot navigation, which comprises the following steps: collecting the geography and physiognomy condition of an indoor and outdoor environment in which a robot is located in real time by a SLAM technology; extracting environmental characteristics; creating a topological map of collected information and establishing the needed different sizes of a topological point according to the posture change of the service robot in the foundation; taking the sizes as input amount to reconstruct the topological map; and generating a topological point self-adapting topological map conforming to the posture change of the mobile robot.

The invention relates to a warehousing system with a simple and convenient positioning function and a cargo delivery method. The system comprises a warehousing cargo shelf, a cargo pallet, a robot forklift and a robot navigation and positioning system, wherein the cargo pallet is mounted on the warehousing cargo shelf; the robot navigation and positioning system comprises a positioning device for the cargo pallet on the cargo shelf and a positioning device for an optional target point on the ground; the robot forklift can optionally walk on the ground in a non-track manner and transversely stores and takes cargoes; the robot forklift comprises a transverse cargo storing and taking device; and one end of the transverse cargo storing and taking device is mounted in a lifting chute in the robot forklift. The warehousing system is reasonable in design, while the robot forklift linearly moves in the front-back direction and the direction is not changed, transverse movement of a sliding plate replaces transverse movement of wheels to longitudinal position the cargo shelf, the problems of transverse deviation rectifying and longitudinal deviation rectifying are solved, and cargo delivery speed is increased effectively.

A mobile, remotely controlled robot includes a turret subsystem, a robot controller subsystem configured to control the robot, control the turret, and fire the weapon, a robot navigation subsystem configured to determine the position of the robot, a turret orientation determination subsystem, and a robot communications subsystem for receiving commands and for transmitting robot position data and turret orientation data. An operator control unit includes a user interface for commanding the robot, the turret, and the weapon. An operator control unit communications subsystem transmits commands to the robot and receives robot position data and turret orientation data from the robot. An operator control unit navigation subsystem is configured to determine the position of the operator control unit. An operator control unit controller subsystem is responsive to the robot position data, the turret orientation data, and the operator control unit position and is configured to determine if the weapon is aimed at the operator control unit within a predetermined fan angle.

The invention provides an indoor robot navigation system and method. The system comprises a man-machine interaction device, a data collection device and a data analysis device; the man-machine interaction device receives terminal-point position coordinate and terminal-point position attitude information input by a user; the data collection device obtains preset position coordinate and present position attitude information, updates a local area map in real time, and obtains a corresponding second coordinate of a dynamic barrier in a 2D global static map via coordinate transformation; and the data analysis device generates a global reference path via a first algorithm, and plans an optimal speed vector of a robot in next time by utilizing a second algorithm to obtain the optimal speed vector, and thus, the robot moves according to the optimal speed vector. The indoor robot navigation system and method are higher in practicality.

The invention discloses a robot navigation method based on color coding identifiers, which comprises the following steps of: extracting a navigation identification line according to the acquired image data, and determining the path direction; identifying a breakpoint mark in front of the color coding identifier, and preparing to regulate the movement trail for the color coding identifier; and identifying a color coding block around a corner or other positions, and executing turning or other command. The invention further discloses a robot navigation system based on the color coding identifiers. The robot navigation system has a simple principle, strong expandability and good visual perception, and can complete the positioning and navigating tasks of an indoor service robot; the robot navigation system can be combined with other navigation systems, so the stability and practicability of the navigation system are improved; and the problems of weaker applicability to complicated environments, error navigation accumulation, large operand, mutual interfered sensors and the like in the current robot navigation field can be effectively solved.

The invention belongs to the technical field of computer vision. A method for semantic segmentation of scene point cloud is provided, a framework of point cloud semantic segmentation model for large-scale dense scene based on depth learning technology is designed, For the input large-scale dense scene point cloud, it can transform the three-dimensional information of point cloud into two-dimensional information which can be processed directly by convolution without losing the information, and complete the task of point cloud semantic segmentation combined with image semantic segmentation technology. In this framework, we can effectively solve the semantic segmentation task of large-scale dense scene point cloud. The semantic segmentation result of the scene point cloud obtained by the method of the invention can be directly used in tasks such as robot navigation, automatic driving and the like. And this method is especially effective in non-synthetic natural scenes.

The invention discloses an indoor robot navigation device which is characterized by comprising a circuit board, wherein a motion control module, a navigation and location module, a map management and path planning module, an image acquisition and processing module, a safety module and a man-machine interaction management module. The invention also provides an indoor robot navigation technology. The technology is characterized by comprising the following steps of: (a) establishing a map and planning a motion path of the robot; (b) refining a area and setting a visual signpost; (c) acquiring and processing the visual signpost, and performing motion location and correction; (d) storing and querying area and signpost information. The device has the advantages of autonomous navigation, high interference resistance capacity, zero track movement and the like, and the problem that most of the conventional navigation modes are low in positioning precision is solved through visual positioning correction.

A method of navigating a robot includes creating a robot navigation map using a map database required for navigation of the robot; and creating a path on which no obstacle is located in the map database using the created robot navigation map. Further, the method of navigating the robot includes primarily controlling the robot so that the robot travels along the created path; and secondarily controlling the robot so that the robot avoids an obstacle on the path.

The invention discloses a double-arm fruit picking robot and a fruit picking method. The double-arm fruit picking robot comprises a fruit picking mechanical device and a control system, wherein the fruit picking mechanical device comprises a left mechanical arm, a right mechanical arm, a left tail end actuator, a right trail end actuator, a moving platform and a fruit basket, the left mechanical arm and the right left mechanical arm are of the same structures, are respectively arranged at the left side and the right side of the moving platform, and respectively complete the work of conveying the left tail end actuator and the right trail end actuator to specified picking points, the left tail end actuator and the right trail end actuator are of the same structures, are respectively arranged at the front end of the left mechanical arm and the front end of the right mechanical arm, and respectively complete the work of obtaining and recovering fruits of fruit trees at the left side and the right of the advancing direction of the robot, and a control system completes the control tasks of robot navigation, walking and picking. The invention also discloses a fruit picking method of the double-arm fruit picking robot. According to the method, the mechanical arms carry out picking in different picking points, and the continuous and respective picking of a plurality of fruits can be realized in one picking point.

The invention provides a robot navigation map construction method based on rat brain hippocampus spatial cells. According to an information transfer loop of spatial navigation related cells in a hippocampus structure of a mammal, a robot acquires a current autonomic movement clue and color depth image information through environment exploration. The autonomic movement clue progressively forms coding to the spatial environment through path integration and characteristic extraction of the spatial cells in the hippocampus structure. The place field of place cells is gradually formed in an exploration process and furthermore covers the whole spatial environment, thereby forming a cognitive map. Meanwhile, Kinect acquires color depth image information of a view scene right in front of the current position as an absolute reference for performing closed-loop detection on the path and correcting a path integration error. At a closed-loop point, a system resets spatial cell discharging activity and performs correction on the path integration error. Nodes on a final navigation map comprise place cell mass coding information, corresponding visual cues and place topological relationships.