A Terrain Sensing Method

Abstract

The invention discloses a terrain perception method, comprising an off-line training part and an on-line classification part. The terrain perception method has the following advantages: 1) a relatively simple and compact feature descriptor and classifier are adopted, thus the consumption of calculation resources and electric quantity can be reduced, the real-time performance of a system is ensured, and a correction algorithm is added to an output back end of the classifier, thereby solving the problem of accuracy reduction; 2) erroneously classified examples are extracted from the classification correction algorithm, and the classifier uses example pairs to perform incremental learning, thereby improving the generalization performance and ensuring the accuracy of long-term running; 3) vibration signals are adopted to perform terrain classification, which is not easily affected by extreme environmental conditions.

Description

technical field [0001] The invention relates to the technical field of robots, in particular to a terrain perception method. Background technique [0002] Wheeled robots will encounter various terrains when moving on the ground, and it is very important to ensure that the robot can safely traverse the terrain and avoid being in a dangerous environment. Existing research mainly focuses on obstacle recognition based on lidar or vision sensors. For impassable obstacles such as walls and stones, we call them "geometric threats". However, the ground itself may also pose a threat to robots. Example: A robot may get stuck in loose sand while traversing it. We call this threat "non-geometric threat". Through real-time terrain perception, the robot can adopt different control strategies according to different terrains, so that it can safely and effectively traverse different terrains. Therefore, real-time terrain perception is very important for wheeled robots working in the fie...

AI Technical Summary

Problems solved by technology

[0004] There are few research results on terrain classification, and the existing work has the following problems: 1) The computing power of the robot is often weak, and it needs a relatively long-lasting battery life

However, many algorithms use complex feature descriptors to represent terrain, which can improve the accuracy of classification to some extent, but it takes up a lot of computing resources, reduces real-time performance, and increases energy consumption; 2) The classifier obtained based on the training data set is effective in a short time, but over time, even samples collected on the same terrain will change greatly

If the parameters of the classifier are not automatically adjusted, the accuracy of the classification will inevitably be reduced.

Images