Summary of Path Following and Stabilisation Of a Bicycle Model Using a Reinforcement Learning Approach, by Sebastian Weyrer et al.
Path Following and Stabilisation of a Bicycle Model using a Reinforcement Learning Approach
by Sebastian Weyrer, Peter Manzl, A. L. Schwab, Johannes Gerstmayr
First submitted to arxiv on: 24 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Robotics (cs.RO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces an innovative Reinforcement Learning (RL) approach to control the motion of a virtual bicycle model, simultaneously achieving path following and lateral stabilization. The Whipple benchmark model is used, with no stabilisation aids, and the agent outputs steering angles that are converted into steering torques via a PD controller. Curriculum learning is employed as the state-of-the-art training strategy, and different settings for the RL framework are investigated and compared. The performance of the deployed agents is evaluated using various paths and measurements, demonstrating their ability to follow complex paths, including full circles, slalom manoeuvres, and lane changes. The paper also uses explanatory methods to analyze the functionality of a deployed agent, linking the introduced RL approach with research in bicycle dynamics. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper teaches machines how to make a virtual bike go along a specific path while keeping stable. This is done using a special kind of machine learning called Reinforcement Learning (RL). The bike is modelled as a complex system and the machine is trained to make the bike follow a desired path, including making turns and changes in direction. The results show that the machines can successfully control the virtual bike at different speeds and on various paths. |
Keywords
» Artificial intelligence » Curriculum learning » Machine learning » Reinforcement learning
