GRASP quadrotor tight-formation paper wins IROS 2026 Best Student Paper
NikolaiMatni · x · 2026-10-01
GRASP Lab's Pei-An Hsieh and Fengjun Yang won the IROS 2026 Best Student Paper Award for "Flatness-Preserving Residual Learning for Real-Time Tight Quadrotor Formation Flight."
- A physics-informed residual dynamics learning framework compensates for turbulent aerodynamic interactions (e.g., downwash) in tight formation flight while preserving differential flatness
- Enables a computationally efficient feedback linearization controller, cutting average tracking error by 31% vs. nominal baselines
- Matches state-of-the-art NMPC tracking with an order of magnitude less computation
- First to show stable tight formation flight with under 30 seconds of training data and a 5ms loop rate
More from Embodied
- Post-acquisition, l2k built a BattleBots robot that's about to get smashed at CoreWeave event — _ScottCondron · 2026-10-01
- Gritt AI robots now install 1 MW of solar per day, 4x faster than human crews — rebeccakaden · 2026-10-01
- Covariant demo shows robot packing items never seen in training data, zero-shot — c_valenzuelab · 2026-10-01
- Unitree T-series humanoid shows off impressive kip-up move — jonstephens85 · 2026-10-01
- Dyna-2.1 Breakdown: How the Team Built the First Physical Agent With 1-Hour Whole-Body Autonomy — chris_j_paxton · 2026-10-01
- Why RL for Low-Level Robot Control: Dyna-2.1 Author's Three Key Reasons — chris_j_paxton · 2026-10-01