Outdoor mobile robot terrain recognition method based on visual contact fusion

Abstract

The invention provides an outdoor mobile robot terrain recognition method based on visual touch fusion. The method comprises the steps that firstly, terrain types are selected according to different road surface materials; tactile and visual modal data are collected for different terrains, and a training set and a test set are divided; a cascade width learning network is constructed, and the network is trained by using the tactile training sample set and the visual training sample set; in the training process, tactile and visual preliminary feature extraction is firstly carried out, then tactile and visual fusion feature extraction is carried out, a fused tactile feature matrix and visual feature matrix are processed through a width learning classification algorithm, and a terrain recognition classification result is obtained through generalized inverse approximation of ridge regression to serve as output of a cascade width learning network; and finally, the test set is input into thetrained cascade width learning network to obtain a terrain recognition classification result. According to the method, visual tactile information is fused based on width learning, terrain recognitionof the mobile robot under small data can be achieved, and robustness and accuracy are high.

Description

technical field [0001] The invention relates to the technical field of robot environment perception, in particular to a terrain recognition method for an outdoor mobile robot based on visual-touch fusion. Background technique [0002] With the rapid development of robot technology, outdoor mobile robots have important applications in tasks such as disaster rescue, field search, and interstellar exploration. In the process of interacting with the environment, the geometric characteristics of the terrain have a great influence on the mobile performance of the robot, and the robot needs to adopt different movement modes for different terrain characteristics. Among them, the flat, hard and high-friction road can move at a relatively high speed, while the soft and muddy road may bring danger to the robot. Therefore, outdoor mobile robots must have fast and effective terrain recognition capabilities. The correct classification of different terrains can greatly improve the autono...

AI Technical Summary

Problems solved by technology

At present, there is no application of width learning algorithm in the field of robot terrain recognition.

Images