The Shortcomings of Force-from-Motion in Robot Learning
Not on the map. Why?
Abstract (arXiv)
Robotic manipulation requires accurate motion and physical interaction control. However, current robot learning approaches focus on motion-centric action spaces that do not explicitly give the policy control over the interaction. In this paper, we discuss the repercussions of this choice and argue for more interaction-explicit action spaces in robot learning.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 14 of its 16 references that are in the map. First posted on arXiv 3 Jul 2024.
Builds on
14 of 16 references.
References · 14
- 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
- 2019Solving Rubik's Cube with a Robot Handcited by 734
- 2020PLAS: Latent Action Space for Offline Reinforcement Learningcited by 133
- 2019Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Taskscited by 67
- 2020Neural Dynamic Policies for End-to-End Sensorimotor Learningcited by 58
- 2023IndustReal: Transferring Contact-Rich Assembly Tasks from Simulation to Realitycited by 53
- 2015From Pixels to Torques: Policy Learning with Deep Dynamical Modelscited by 45
- 2019Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assemblycited by 39
- 2021LASER: Learning a Latent Action Space for Efficient Reinforcement Learningcited by 32
- 2023On the Role of the Action Space in Robot Manipulation Learning and Sim-to-Real Transfercited by 26
- 2020Learning Vision-based Reactive Policies for Obstacle Avoidancecited by 12
- 2021Learning to Centralize Dual-Arm Assemblycited by 9
Built on by
Citing papers per year: 2026: 1.
1 paper.