Gunji Sugimoto

An Automatic Guidance System of a Self-Controlled Vehicle

An automatically guided vehicle, traveling without fixed guide ways, has been developed. In this paper, the construction of the vehicle, the control algorithm, and its general performance are described.

Neural network control for automatic braking control system

Abstract We have developed an automatic braking control system for automobiles applying a three-layer neural network model. This system enables the vehicle to decelerate smoothly and to stop at the specified position behind the vehicle ahead. This paper focuses on the use of the three-layer neural network model in the automatic braking control system. Because vehicle dynamics is varied by the variations in road conditions or vehicle characteristics, it cannot be represented accurately by mathematical models. According to this reason, the conventional control methods, such as proportional integrative derivative (PID) control, cannot achieve satisfactory control performance. Therefore, we have constructed the neural network adaptive control system based on the feedback error learning method. This learning method enables the system to adapt to the changes in road grade and vehicle weight without using any specific sensors. Experimental results show satisfactory control performance and reveal that the neural network adaptive control system based on the feedback error learning method is available for the automatic braking control system.

Evolution of machine intelligence for mobile robots

Active research and development is carried out for various mobile robots or robotic vehicles. This paper tries to find a new principle of the evolution of machine intelligence for robotic vehicles. The machine intelligence of robotic vehicles is the integrated ability composed of sensors, actuators, and computers, which produces the travel versatility of robots. Therefore, the relation between the performances of sensors, actuators, and computers, and the produced travel versatility was examined on 18 previously reported wheeled robotic vehicles with reference to the study on the evolution of vertebrate brains. The obtained hypothetical principles are as follows. Computer power varies in proportion to the summation of both rates of information input from sensors and information output to actuators for robots with the same grade of travel versatility. Robotic vehicles with a high grade of travel versatility have greater computer power than those with a low grade.