Guang-Yi Cao

Reinforcement Learning Neural Network to the Problem of Autonomous Mobile Robot Obstacle Avoidance

An approach to the problem of autonomous mobile robot obstacle avoidance using reinforcement learning neural network is proposed in this paper. Q-learning is one kind of reinforcement learning method that is similar to dynamic programming and the neural network has a powerful ability to store the values. We integrate these two methods with the aim to ensure autonomous robot behavior in complicated unpredictable environment. The simulation results show that the simulated robot using the reinforcement learning neural network can enhance its learning ability obviously and can finish the given task in a complex environment.

The effect of the fractional-order controller's orders variation on the fractional-order control systems

This paper concerns fractional-order controllers. The definitions and properties of fractional calculus are introduced. The mathematical descriptions of the fractional-order system and the fractional-order controller are outlined. The variation of the fractional-order controllers order and its effect on the fractional-order control systems are investigated by qualitative analysis and simulation study. Conclusions and simulation examples are given.

Predictive control of proton exchange membrane fuel cell (PEMFC) based on support vector regression machine

A new method of the predictive control for proton exchange membrane fuel cell (PEMFC) based on support vector regression machine is presented and the support vector regression machine is constructed. The process plant is modeled on SVRM. The predictive control law is obtained by using the particle swarm optimization (PSO).The simulation and the results show that the support vector regression machine and the PSO receding optimization applied to the PEMFC predictive control have good performance.

Described Model of a Modular Self-Reconfigurable Robot

A homogeneous lattice modular Self-Reconfigurable (MSR) robot was designed in the paper, we discuss how to describe configuration of robot using graphs theory, how to discover a robot configuration, utilizing connecting status of six connection component, a feature vector matrix was proposed in order to accurately described the topology structure, position and connection relation of MSR robot. Basic movement and meta-module structure was introduced too. A locomotion example of robot was demonstrated, it verify the correctness of MSR robot model described. The methodology establishes foundation to further study self-reconfiguration algorithm, it is very general and can be applied easily to other modular robots.