Learning to ride a bicycle with reinforcement learning

This is an Online simulation using matlab, where the agent learns to ride a bicycle using SARSA (lambda) algorithm (reinforcement learning). This was inspired from the paper written by J. Randlov and P. Alstrom.

The bicycle model was acquired from http://model.3dcool.net/ and then modified using Autodesk 3D studio Max (student version).

This work was done as a part of the summer research project in Information: Signals, Images, Systems (ISIS) Research Group, which is part of the School of Electronics and Computer Science at the University of Southampton. The project was supervised by Porf. Mahesan Niranjan (Head of ISIS) and conducted by Alma A. M. Rahat (pt. III student in ECS).

More details about this project could be found in www.almarahat.site11.com !

Category:

Science & Technology

Tags:

• reinforcement
• learning;
• machine
• SARSA;
• SARSA(lambda);
• Bicycle
• animation;
• 3D
• online
• bicycle;
• control;
• learning
• Matlab;
• animation
• using
• matlab;
• matlab

[almarahat]

@tashrifahmed Thanks ! :-D

[tashrifahmed]

nice job Rahat.. Thumbs up!