Autonomous Obstacle Removal for Excavators through Policy Learning with Particle Simulation
2026-06-08 • Robotics
Robotics
AI summaryⓘ
The authors tackled the challenge of teaching a robotic excavator how to remove buried obstacles efficiently. They created a simulation that mimics real digging conditions using particles, but since deep obstacles need complex simulations, they developed a step-by-step learning plan starting with easy, shallow obstacles and gradually increasing difficulty. This approach made training much faster and allowed the learned skills to work well on a real excavator dealing with different buried objects. Their method outperformed other attempts that took much longer without success.
autonomous excavationparticle-based simulationcurriculum learningsim-to-real transfersoil-obstacle interactionexcavation trajectoryRGB-D sensingterrain deformationrobotic manipulation
Authors
Yuki Kadokawa, Sandro M. Alcantara Tacora, Taro Abe, Daisuke Endo, Genki Yamauchi, Takeshi Hashimoto, Takamitsu Matsubara
Abstract
Autonomous obstacle removal from the ground is an important earthwork task, but this is difficult to automate because an excavator must adapt its excavation trajectories over repeated cycles as soil-obstacle conditions change. Learning such state-dependent behavior requires a training environment that reproduces accumulated soil-obstacle interactions, including contact states, terrain deformation, and obstacle visibility. Accordingly, particle-based simulation is suitable for the relevant policy learning. However, particle simulation is computationally expensive, and repeated excavation cycles further increase the learning cost. We observe that the burial condition of an obstacle governs both task difficulty and simulation cost: deeper burial makes obstacle removal harder while also requiring more particles for accurate simulation. This observation motivates a burial-conditioned curriculum learning strategy. We propose a time-efficient sim-to-real policy learning framework in which the policy observes terrain and obstacle information from RGB-D measurements and then outputs a parameterized excavation trajectory; in this process, the simulator reproduces in a real-world excavator the same observation-action interface it uses under controllable burial conditions. The curriculum begins with shallow burial conditions and progressively increases burial depth while adjusting particle count, thus simultaneously controlling task difficulty and simulation cost. Experiments show that the proposed framework successfully learns an effective obstacle-removal policy, whereas baseline methods fail even after a full week of training. The proposed curriculum achieves effective performance within three days and achieves successful transfer to a real 12-ton excavator operating on open ground with various steel obstacles, thus demonstrating robust obstacle removal.