Koshien robot team shares control stack: nonlinear-opt trajectories hit ±1.2cm average error
4310sy · x · 2026-11-16
- Trajectory generation: nonlinear optimization precomputes time-minimal trajectories, 0.1s per solve, 50 trajectories in 3s; motor acceleration/speed limits are constraints so trajectories are physically trackable; robot approximated as a square for obstacle clearance that scales with speed; switched from IPOPT to FATROP for faster solving
- Trajectory tracking: nearest-point tracking with a computed acceleration-to-current FF coefficient (avoiding D-gain tuning), steering motor velocity FF, and a velocity-vector rotation correction proportional to turning rate
- Localization: steering wheel odometry + MCL with three LiDARs, addressing timestamp misalignment in point cloud merging
- Result: ±1.2cm average position error start to finish; code to be open-sourced in natto library
More from Embodied
- microVLA runs pi0.5 VLA training and inference on a 4GB RTX 3050, 2.5x latency cut — this_is_surabhi · 2026-11-16
- Most VLAs lose their VLM's vision-language skills during pretraining — DJiafei · 2026-11-16
- RL environments become a business: Mercor at $20B valuation, Lumos launches long-horizon robot RL envs — 量子位 · 2026-11-16
- Developer's New Robotics BLDC Controller PCB Batch Arrives, Running Unit Tests — wavefnx · 2026-11-16
- YC-backed Ornadyne's flapping-wing drone was classified as a bird by Air Force detectors — import_jmr · 2026-11-16
- Uber starts offering Amazon's Zoox robotaxi rides in Las Vegas — Polymarket · 2026-11-16