Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
Tham khảo tài liệu 'robotics automation and control 2011 part 9', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Genetic Reinforcement Learning Algorithms for On-line Fuzzy Inference System Tuning Application to Mobile Robotic 231 0.1 S U I 4 uL a S t 1 t 1 t 1 t 1 t 1 R t 1 lt-1 This value will be used to calculate the TD error in the next step time. However the performances of the controller are closely dependent on the correct choice of the discrete actions set witch is determined using a priori knowledge about system for complex systems like robots priori knowledge are not available then it becomes difficult to determine a set of correct actions in which figure the optimal action for each fuzzy rule. To solve this problem and to improve the performances of the reinforcement learning the genetic algorithms will explore the broadest space of solutions to find the solution optimal Dongbing Gu Huosheng Hu Libor Spacek Chia-Feng Juang Min-Soeng Kim and Kim and Ju-Jang Lee Chia-Feng Juang and Chun-Feng Lu and that without any priori knowledge. 4. Genetique reinforcement algorithmes Genetic Algorithms GA are stochastic optimization algorithms founded on species evolution mechanisms David E Goldberg. In GA a candidate solution for a specific problem is called an individual or a chromosome and consists of a linear list of genes. Each individual represents a point in the search space and hence a possible solution to the problem. A population consists of a finite number of individuals. Each individual is decided by an evaluating mechanism to obtain its fitness value. Based on this fitness value and undergoing genetic operators a new population is generated iteratively with each successive population referred to as a generation. Genetic Reinforcement Learning enable to determine the best set of parameters of antecedents consequences starting from a random initialization of these parameters and in each iteration only one action is applied on the system to the difference of a traditional genetic algorithm GA use three basic operators the selection crossover and mutation to manipulate .
