In this paper a learning mechanism for reactive fuzzy controller design of a mobile robot navigating in unknown environments is proposed. The fuzzy logical controller is constructed based on the kinematics model of a ...In this paper a learning mechanism for reactive fuzzy controller design of a mobile robot navigating in unknown environments is proposed. The fuzzy logical controller is constructed based on the kinematics model of a real robot. The approach to learning the fuzzy rule base by relatively simple and less computational Q-learning is described in detail. After analyzing the credit assignment problem caused by the rules collision, a remedy is presented. Furthermore, time-varying parameters are used to increase the learning speed. Simulation results prove the mechanism can learn fuzzy navigation rules successfully only using scalar reinforcement signal and the rule base learned is proved to be correct and feasible on real robot platforms.展开更多
Motion simulator usually appears the phenomenon of false cues and the workspace is limited in the process of washout. The proposed washout algorithm combines fuzzy logic control with the vestibular system to design th...Motion simulator usually appears the phenomenon of false cues and the workspace is limited in the process of washout. The proposed washout algorithm combines fuzzy logic control with the vestibular system to design the tilt coordination fuzzy adaptive filter, in order to minimize the vestibular sensory error below the human perception threshold. Owing to tilt coordination angular velocity limiter, the loss of low-pass acceleration must be compensated by the acceleration transform model. The translational channel decreases the possibility of the workspace beyond limitation and expands the scope of motion platform simulating input acceleration by using third-order filter. The simulation results show that the proposed algorithm can effectively overcome the phase retardation of classical washout algorithm, and then prevent the produce of false cues, decrease the displacement of motion platform simultaneously; in addition, white Gaussian noise simulates large variations in acceleration. The proposed washout algorithm can have maximal extreme value of acceleration and accurate simulating performance in general. It also proves that the proposed washout algorithm has a strong adaptability and reliability, which can effectively improve the dynamic fidelity for motion simulator.展开更多
文摘In this paper a learning mechanism for reactive fuzzy controller design of a mobile robot navigating in unknown environments is proposed. The fuzzy logical controller is constructed based on the kinematics model of a real robot. The approach to learning the fuzzy rule base by relatively simple and less computational Q-learning is described in detail. After analyzing the credit assignment problem caused by the rules collision, a remedy is presented. Furthermore, time-varying parameters are used to increase the learning speed. Simulation results prove the mechanism can learn fuzzy navigation rules successfully only using scalar reinforcement signal and the rule base learned is proved to be correct and feasible on real robot platforms.
基金Funded by the National Natural Science Foundation of China(U1233107)Civil Aviation Science and Technology Innovation Project of China(MHRD20140210)
文摘Motion simulator usually appears the phenomenon of false cues and the workspace is limited in the process of washout. The proposed washout algorithm combines fuzzy logic control with the vestibular system to design the tilt coordination fuzzy adaptive filter, in order to minimize the vestibular sensory error below the human perception threshold. Owing to tilt coordination angular velocity limiter, the loss of low-pass acceleration must be compensated by the acceleration transform model. The translational channel decreases the possibility of the workspace beyond limitation and expands the scope of motion platform simulating input acceleration by using third-order filter. The simulation results show that the proposed algorithm can effectively overcome the phase retardation of classical washout algorithm, and then prevent the produce of false cues, decrease the displacement of motion platform simultaneously; in addition, white Gaussian noise simulates large variations in acceleration. The proposed washout algorithm can have maximal extreme value of acceleration and accurate simulating performance in general. It also proves that the proposed washout algorithm has a strong adaptability and reliability, which can effectively improve the dynamic fidelity for motion simulator.
