Amicro Semiconductor Co., Ltd.

Other

Amicro Semiconductor Corporation

Gangjun Xiao

Yongyong Li

Qinwei Lai

Taiming Huang

Ming Li

Dengke Xu

Jianfeng Dai

Weibing Zhao

Yongxian Yan

Gentang Li

Wang Zhao

The disclosure discloses a method for building a local point cloud map and a visual robot. According to the method for building the local point cloud map, point clouds which reflect a position of a large-range region around a robot are sampled according to a preset salient pose change condition; point clouds at different height positions are described in the form of a 3d histogram.

Method for building a local point cloud map and vision robot

A recharging control method of a desktop robot is provided. The desktop robot includes a robot body provided with a camera and a base. The recharging control method includes the following steps. When the robot body of the desktop robot receives a charging instruction, images of the desktop environment are collected, the collected two continuous frames of images are controlled to be subjected to feature matching, the position variation of the robot body is calculated, and then a corresponding grid map is constructed (step A); a wireless beacon modulated and transmitted by the base is received, and initial orientation of the base is analyzed according to strength information of the wireless beacon (step B); a recharging path is planned in combination with the change of the position of the robot body and the initial orientation obtained through analysis (step C); the robot body of the desktop robot is controlled to move along the recharging path (step D); when the strength of the wireless beacon is within a preset threshold range, the collected two continuous frames of images are controlled to be subjected to feature matching, then a vector value of the current position of the robot body is calculated, and further, the base is further searched for completing docking and charging (step E).

Recharging control method of desktop robot

The present disclosure relates to a method for detecting a skidding of a robot, a mapping method and a chip. A method for detecting a skidding of a robot, comprising the following steps: By an odometer on existing driving wheels of a robot and a gyroscope and a processor in a body of the robot, a first angle change rate generated by two driving wheels within a preset time and a second angle change rate generated by the gyroscope within the preset time are detected and calculated, so as to determine an angular velocity change error rate of the robot. Finally, by judging whether the angular velocity change error rate is greater than or equal to a preset value, it is determined whether the robot is in the skidding state.

Method for detecting skidding of robot, mapping method and chip

A robot repositioning method is provided. A position deviation caused by excessive accumulation of a walking error of a robot may be corrected to implement repositioning by taking a path that the robot walks along an edge of an isolated object as a reference, so that the positioning accuracy and walking efficiency of the robot during subsequent navigation and walking are improved.

Method for robot repositioning

The disclosure relates to a method for predicting and controlling robot walking. The method includes the following steps: constructing a grid map based on grid units marked with a status; establishing a dynamic detection model with a current location of a robot as a reference point based on the grid map; predicting a forward path condition of the robot based on the dynamic detection model; and controlling a walking mode of the robot based on the prediction result.

The disclosure relates to a method for monitoring a pet by a robot based on a grid map and a chip. A mutual position relationship between the pet and the robot is determined through wireless communication between a wireless signal device on the pet and the robot, and then whether there is an obstacle cell or not between grid cells where the robot and the pet are located in the grid map is judged. If NO, it is indicated that the pet may be effectively shot at a present position and shooting direction of the robot, and the present position and shooting direction of the robot are not required to be changed. If YES, it is indicated that no pet but an obstacle may be shot at the present position of the robot.

Method for monitoring pet by robot based on grid map and chip

The present disclosure provides an intelligent pet monitoring method for a robot. A mutual position relationship between a pet and a robot is determined according to wireless communication between a wireless signal device on the pet and the robot, and then a grid position corresponding to the pet in a grid map constructed by the robot is determined according to the mutual position relationship. Daily activities of the pet may also be regular, to a certain extent, under the influence of a biological clock of the pet, so that big data generated by the grid positions of the pet at different time points is finally recorded to construct the pet daily activity model, and the daily activity trajectory of the pet may be predicted according to the pet daily activity model.

Intelligent pet monitoring method for robot

A method for controlling a motion of a robot based on map prediction mainly carries out estimation and calculation for a wall surface by combining an external sensor with internal map information about a robot, so as at least to enable the robot to walk along the estimated wall surface. The method for controlling the motion of the robot based on map prediction can be adapted to various different wall surfaces based on map prediction, including different colors and shapes, thereby reducing an operation time; and the accuracy of map prediction can be continuously corrected during an operation process, thereby realizing a good wall-following behavior.

Method for controlling motion of robot based on map prediction

A path planning method of an intelligent robot includes: (1) dividing the operating space into a plurality of rectangular virtual regions having areas of n*m square meters; (2) controlling the intelligent robot to traverse each rectangular virtual region with bow-shaped behavior; (3) when moving in bow-shape, building a raster map with the distance information acquired by distance information sensor, angle information acquired by angle information sensor, collision information acquired by collision detection sensor and obstacle information acquired by obstacle detection sensor, and marking the pass points, obstacle points and collision points; (4) analyzing the raster map in real time to judge if the entire operating space has been fully traversed: if so, moving to step (5); otherwise, returning to step (2); (5) controlling the intelligent robot to make an alongst behavior, and marking the alongst behavior points on the raster map.

Path planning method of intelligent robot

A method of planning a cleaning route for a cleaning robot: firstly, starting from an original point based on maps of grids, cleaning grid zones formed by the grids one by one until an entire region is cleaned, and then establishing a map of the entire region; secondly, searching the map of the entire region to find out uncleaned areas missed from cleaning, and then cleaning the uncleaned areas; thirdly, cleaning peripheral areas of the entire region based on the map of the entire region; and lastly, returning to the original point. A chip is also provided which stores procedures for controlling the cleaning robot to implement the method of planning a cleaning route.

Growing community of inventors

Guangdong, China

Gangjun Xiao