Description: 基于BP神经网络的 参数自学习控制
（1）确定BP网络的结构，即确定输入层节点数M和隐含层节点数Q，并给出各层加权系数的初值 和 ，选定学习速率 和惯性系数 ，此时k=1；
（2）采样得到rin(k)和yout(k)，计算该时刻误差error(k)=rin(k)-yout(k)；
（3）计算神经网络NN各层神经元的输入、输出，NN输出层的输出即为PID控制器的三个可调参数 ， ， ；
（4）根据(3.34)计算PID控制器的输出u(k)；
（5）进行神经网络学习，在线调整加权系数 和 ，实现PID控制参数的自适应调整；
（6）置k=k+1，返回（1）。
-Based on the parameters of BP neural network self-learning control (1) to determine the structure of BP network, that is, determine the input layer nodes M and hidden layer nodes Q, and gives all levels of the initial value and the weighted coefficient, the selected learning rate and inertia coefficient, when k = 1 (2) sample has been rin (k) and the yout (k), calculate the moment of error error (k) = rin (k)-yout (k) (3) calculation of neural network NN all floors of the neurons in input and output, NN output layer is the output of PID controller for the three adjustable parameters,, (4) According to (3.34) Calculation of PID controller output u (k) (5) to carry out neural network learning, on-line adjustment of the weighted coefficient and, realize the adaptive PID control parameters adjust (6) purchase k = k+ 1, return (1). Platform: |
Size: 1024 |
Author:dake |
Hits:
Description: Simulink 控制VR环境中的小车。小车有5个距离传感器，能够慢慢学会避开墙壁和障碍物。小车采用加强学习（Q learning），采用神经网络对Q函数逼近。由于使用了模拟退火，小车在开始的时候会经常撞击障碍物，10次后基本就不会再撞了。
小车的外观模型使用了"w198406141"在本论坛的虚拟现实区发布的VR模型。-VR environment Simulink control car. There are 5 car distance sensor, can gradually learn to avoid walls and obstacles. Car used to enhance learning (Q learning), using neural networks Q function approximation. As the use of simulated annealing, the car will start when the regular crash barrier, 10 times the fundamental will no longer hit. The appearance model used car " w198406141" in this forum area of virtual reality VR model release. Platform: |
Size: 12288 |
Author:gao |
Hits:
Description: State Space Q-Learningfor control of nonlinear system- State Space Q-Learningfor control of nonlinear system Platform: |
Size: 1349632 |
Author:周彦一 |
Hits:
Description: Simulink 控制VR环境中的小车。小车有5个距离传感器，能够慢慢学会避开墙壁和障碍物。小车采用加强学习（Q learning），采用神经网络对Q函数逼近。由于使用了模拟退火，小车在开始的时候会经常撞击障碍物，10次后基本就不会再撞了。
-VR environment Simulink control car. There are 5 car distance sensor, can gradually learn to avoid walls and obstacles. Car used to enhance learning (Q learning), using neural network Q function approximation. As the use of simulated annealing, the car in the beginning often hit obstacles, and then ,10 times later, it shouldn t happen. Platform: |
Size: 179200 |
Author:zhangziyang |
Hits:
Description: Reinforcement Learning 中以 Q learning 學習的倒單擺實驗。以x,x_dot,theta,theta_dot作為狀態參數 state，利用model模擬之結果做Q值更新，產生Q table，對學習平衡控制。結果以csv檔作為輸出，可以由使用者作圖觀察變化。-Reinforcement Learning , utilize Q learning for An inverted pendulum system control。x,x_dot,theta,theta_dot as state variable，use model result renew Q table，control learning of balance pole 。output csv file contain trail(times),balance time(sec).it s better observe the change if user plot it. Platform: |
Size: 316416 |
Author:王曉明 |
Hits:
Description: This package provides a Lasagne/Theano-based implementation of the deep Q-learning algorithm described in:
Playing Atari with Deep Reinforcement Learning Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra, Martin Riedmiller
and
Mnih, Volodymyr, et al. "Human-level control through deep reinforcement learning." Nature 518.7540 (2015): 529-533.
Here is a video showing a trained network playing breakout (using an earlier version of the code):
http://youtu.be/SZ88F82KLX4 Platform: |
Size: 26624 |
Author:YH.HO |
Hits:
Search - q learning control