aiqbench analysis
This research introduces a computationally efficient sim-to-real framework for autonomous excavators that leverages burial-conditioned curriculum learning to balance task difficulty with simulation cost. By dynamically adjusting particle counts based on obstacle depth, the method achieves successful transfer to real-world 12-ton excavators in three days, significantly outperforming baseline methods that require weeks of training. The approach demonstrates a practical solution for automating complex earthwork tasks where state-dependent behaviors must adapt to changing soil-obstacle interactions.
Autonomous Obstacle Removal for Excavators through Policy Learning with Particle Simulation
arXiv:2606.09183v1 Announce Type: new 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.
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.
Robotics (cs.RO)
arXiv:2606.09183 [cs.RO]
(or arXiv:2606.09183v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.09183 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 334 words
Related coverage
From Target Selection to Digging: A Learning-Based Framework for Continuous Autonomous Excavation
arXiv · Sep 25, 2026
Vision Guided Target Conditioned Control for Autonomous Excavation
arXiv · Aug 25, 2026
This startup’s self-driving excavators help dig up $600M in new funding
TechCrunch · Oct 26, 2026
ExcavaTwin: Training-Free Geometry-Guided Semantic Elevation Mapping for Autonomous Excavation
arXiv · Sep 29, 2026