Showing 2025Show all
2 papers · 1 filter
cs.RO2025
StableTracker: Learning to Stably Track Target via Differentiable Simulation
Fanxing Li, Shengyang Wang, Fangyu Sun +5
Existing FPV object tracking methods heavily rely on handcrafted modular pipelines, which incur high onboard computation and cumulative errors. While learning-based approaches have…
cs.RO2025
ABPT: Amended Backpropagation through Time with Partially Differentiable Rewards
Fanxing Li, Fangyu Sun, Tianbao Zhang +1
Quadrotor control policies can be trained with high performance using the exact gradients of the rewards to directly optimize policy parameters via backpropagation-through-time (BP…