Parking Motion Planning and Control of a Car-Like Robot Using a Fuzzy Neural Network
Motoji Yamamoto, Masaaki Kobayashi, and Akira Mohri
Dept. of Intelligent Machinery and Systems, Faculty of Engineering, Kyushu University, 6-10-1 Hakozaki, Higashi-ku, Fukuoka, 812 Japan
This paper discusses a parking motion planning and control of a car-like robot. Because of non-holonomic constraints of the system, motion planning and control is regarded as a difficult problem. In this paper, constraints of steering operation and obstacle avoidance with garage and walls are also considered. As one approach to this problem, extracting human control strategy can be considered, because many drivers can easily park their cars in garages. This paper proposes a motion planning and control method using a fuzzy neural network (FNN). The fuzzy neural network system for parking motion planning learns good parking motions by human operations to generate motion strategy of parking. The fuzzy neural network is then used for parking motion planning in a restricted area surrounded by walls. Computer simulation demonstrates the effectiveness of the planning method. Furthermore, the method can be considered as a feedback control law for the parking of car-like robot. Therefore, an experiment of parking motion control using the fuzzy neural network is also tested.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 International License.