{"ID":2823682,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.24740","arxiv_id":"2512.24740","title":"Control of Microrobots with Reinforcement Learning under On-Device Compute Constraints","abstract":"An important function of autonomous microrobots is the ability to perform robust movement over terrain. This paper explores an edge ML approach to microrobot locomotion, allowing for on-device, lower latency control under compute, memory, and power constraints. This paper explores the locomotion of a sub-centimeter quadrupedal microrobot via reinforcement learning (RL) and deploys the resulting controller on an ultra-small system-on-chip (SoC), SC$μ$M-3C, featuring an ARM Cortex-M0 microcontroller running at 5 MHz. We train a compact FP32 multilayer perceptron (MLP) policy with two hidden layers ($[128, 64]$) in a massively parallel GPU simulation and enhance robustness by utilizing domain randomization over simulation parameters. We then study integer (Int8) quantization (per-tensor and per-feature) to allow for higher inference update rates on our resource-limited hardware, and we connect hardware power budgets to achievable update frequency via a cycles-per-update model for inference on our Cortex-M0. We propose a resource-aware gait scheduling viewpoint: given a device power budget, we can select the gait mode (trot/intermediate/gallop) that maximizes expected RL reward at a corresponding feasible update frequency. Finally, we deploy our MLP policy on a real-world large-scale robot on uneven terrain, qualitatively noting that domain-randomized training can improve out-of-distribution stability. We do not claim real-world large-robot empirical zero-shot transfer in this work.","short_abstract":"An important function of autonomous microrobots is the ability to perform robust movement over terrain. This paper explores an edge ML approach to microrobot locomotion, allowing for on-device, lower latency control under compute, memory, and power constraints. This paper explores the locomotion of a sub-centimeter qua...","url_abs":"https://arxiv.org/abs/2512.24740","url_pdf":"https://arxiv.org/pdf/2512.24740v1","authors":"[\"Yichen Liu\",\"Kesava Viswanadha\",\"Zhongyu Li\",\"Nelson Lojo\",\"Kristofer S. J. Pister\"]","published":"2025-12-31T09:18:36Z","proceeding":"cs.RO","tasks":"[\"cs.RO\",\"eess.SY\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}