โ Scribed by Ranajit Chatterjee; Fumitoshi Matsuno
No coin nor oath required. For personal study only.
In an effort to mimic the process of obstacle avoidance behavior of human locomotion (or that of automobiles driven by human response), a combination of basic reflex actions and higher level logical decisions is implemented. It is shown that for reflective navigation of autonomous mobile robots, the ability to reflectively avoid obstacles on one side (left or right) only is sufficient for avoidance of obstacle on both sides. When combined with a free-target-approach behavior, the robot can be made capable of navigating through environments with unknown obstacles towards a desired target. Use of this behavior segmentation provides a basis for a compact representation of the reflex behavior as well as allows a concept of virtual-target-side to solve the limit cycle problems, which is common in reflective navigation. In the present work, fuzzy logic based implementation of the single-sided reflex is considered. The use of perceptional symmetry allows perception-action mapping with reduced sensor space dimensions. Simulation and experimental results are presented to show the effectiveness of the proposed strategy in typical obstacle situations.
๐ SIMILAR VOLUMES
A neuro-fuzzy system architecture for behavior-based control of a mobile robot in unknown environments is presented. A neural network is used to understand environments. Its inputs are a heading angle between the robot and a specified target, and range information acquired by an array of ultrasonic
Here, the authors present new results regarding the problem of driving an autonomous vehicle by the use of vision. The paper deals with three aspects of the system: the vision algorithm, the vehicle's behaviour, and the lateral control, making the guidance possible. The main point of this paper is t