Reproduce “Bridging the Gap Between Safety and Real-Time Performance in Receding-Horizon Trajectory Design for Mobile Robots” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shreyas Kousik, Sean Vaskov, Fan Bu, Matthew Johnson-Roberson, Ram Vasudevan
Researchers developed the Reachability-based Trajectory Design (RTD) method, which generates provably safe and persistently feasible trajectories for mobile robots operating in real-time within uncertain environments. This approach achieved 0 crashes in extensive simulations and demonstrated safe operation on physical robot platforms, overcoming computational challenges for complex robot dynamics.
Key insights
- Decomposing a high-dimensional robot system into lower-dimensional self-contained subsystems allows for practical FRS computation using SOS programming, significantly reducing memory consumption (e.g., from >3.5 TB to 473 GB for a Rover model).
- A discrete, buffered representation of obstacles enables real-time evaluation of collision avoidance constraints within an online optimization framework without compromising the provable safety guarantees provided by FRS.
- Explicitly bounding and incorporating tracking and state estimation errors into the FRS definition and obstacle buffering is essential for robust safety guarantees in real-world operational environments with inherent uncertainties.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.