Estimation of Distance, Sizes and Angles of Obstacles for Fish Robots

摘要

We designed and implemented fish robots for various purposes such as autonomous navigation, maneuverability control, posture balancing and improvement of quick turns in a tank. Typically, fish robots have 30-50 X 15-25 X 10-20cm dimensions; length, width and height, respectively. Obstacle detection and recognition to avoid collision for a natural and smooth movement are the most important concerns for a fish robot. Walls, rocks, water plants and other nearby robots are common obstacles for a group of small scale fish robots and submersibles that have been constructed in our lab. A distance measuring sensor for general purpose use rather than a camera is mounted on a motor shaft to scan a certain range of foreground from the head of a fish robot. The fish robot's ability to recognize the distance, size, and the approaching angles to an obstacle is improved to avoid collision based on the scanned information. Evident features of obstacles sizes and angles are obtained through a neural network training algorithm from the scanned data by a simple distance sensor. Therefore, a scanning distance sensor module can replace a camera which is necessary to get front information about obstacles. Experimental results show the successful path control of a fish robot to go through a passage without hitting obstacles.