Demo

Features

MultiTaskOneArmerAgent inspired by Example Guided Deep Reinforcement learning Deepmimic This white reinforcement learned agent satisfy immitation of yellow referenceground truth motion)

Immitation Objective
Additional task objective Be able to choose below 7 Different Mode with your gui. ・4 Independently Trained Task +WaitMode+Manual Mode with keyboard input+dangerous(not trained enough) Mode.

Documentation

For more information, in addition to installation and usage instructions, see my website below.
Performance & Concept https://qiita.com/tanakataiki/items/94c7778341a40438cc99
Enviroment Setting https://qiita.com/tanakataiki/items/40f54b3851c2ed3e09ec
Unity C# https://qiita.com/tanakataiki/items/fbb3a4782add01d73428
Tensorflow Search https://qiita.com/tanakataiki/private/8655c76b68933905f3ea
If you are a researcher interested in a discussion of this project let me know !
If you have used a version of the ML-Agents toolkit prior to v0.5, we strongly recommend our guide on migrating from earlier versions.

License

Also This Repo Includes Unity ML-Agent License Apache License 2.0

Citation

If you use MultiTask Armer Agent to conduct research, we ask that you cite this website as a reference.

If you use Unity or the ML-Agents Toolkit to conduct research, we ask that you cite the following paper as a reference:

Juliani, A., Berges, V., Vckay, E., Gao, Y., Henry, H., Mattar, M., Lange, D. (2018). Unity: A General Platform for Intelligent Agents. arXiv preprint arXiv:1809.02627. https://github.com/Unity-Technologies/ml-agents.

Reference

Paper:https://xbpeng.github.io/projects/DeepMimic/2018_TOG_DeepMimic.pdf
Blog : https://bair.berkeley.edu/blog/2018/04/10/virtual-stuntman/
Code : https://github.com/xbpeng/DeepMimic

Acknowledgement

This reinforcement learning projet is under the support of SE4 inc

Thanks for Loki Savkin and Andri

YAD2U

Install / Use

README