Hi everyone,
I hope you are well today,
OK, let me go straight forward, I really like to implement the new control techniques, but I'm having to much problem in the concepts, it will be really helpful if I can see some codes about RL in Control Theory.
So I will be really thankful if you can share me some codes, whether you wrote it or you see them somewhere else.
Thanks a lot,
Sincerely,
Arash.
Hi Arash,
This code for Reinforcement Learning applied to Linear Quadratic Regulator , can help you https://github.com/papachristoumarios/ntua-neuro-fuzzy-control
Dear Arash Marashian,
I suggest you to see links and attached files on topic.
http://www.wildml.com/2016/10/learning-reinforcement-learning/
https://towardsdatascience.com/applications-of-reinforcement-learning-in-real-world-1a94955bcd12
https://orbi.uliege.be/bitstream/2268/27963/1/book-FA-RL-DP.pdf
https://medium.com/datadriveninvestor/training-a-robotic-arm-to-do-human-like-tasks-using-rl-8d3106c87aaf
https://arxiv.org/pdf/2002.09820.pdf
Article Emotion in Reinforcement Learning Agents and Robots: A Survey
Best regards
Hi Arash,
In our recent paper (https://arxiv.org/abs/2006.03116), we discuss Policy Gradient approaches to learn the optimal LQR control for Markovian Jump Linear Systems, but the results can easily be applied to LTI systems.
We present an algorithm to estimate the gradient of the cost function, which can be used to improve the controller gains, and iterate towards the optimal LQR gains (this is similar to REINFORCE).
The code used in our simulations is provided here: https://github.com/jpjporto/MJLS_Learning
Here, we use a Monte Carlo algorithm to estimate the gradient, but that step can also be replaced by common RL libraries like Tensoflow.
Best,
Joao
Hi Arash,
Check out Reinforcement Learning Toolbox User’s Guide by Matlab. There is a lot of RL implementation example code.
Regards,
Hi Arash,
This code for Reinforcement Learning applied to Linear Quadratic Regulator , can help you https://github.com/papachristoumarios/ntua-neuro-fuzzy-control
Thanks El Hajjami Lhoussain , I read the related paper and the code, but I think there might be a problem in the code. If you have time we can talk about it later.
sincerely,
A. Marashian
Hi Arash Marashian
The code is based on a NIPS paper "Reinforcement Learning applied to Linear Quadratic Regulator".https://papers.nips.cc/paper/712-reinforcement-learning-applied-to-linear-quadratic-regulation.pdf
In my opinion I believe that this code presents several problems .
sincerely,
the stable-baselines library contains python codes for many of the current RL algorithms, and is easy to run with python:
https://github.com/hill-a/stable-baselines
- Badges
- Science topic
- Similar topics
- Engineering
- Control Theory