Continuous control actions learning and adaptation for robotic manipulation through reinforcement learning
Problem
Programming robots to perform continuous-control object manipulation tasks and adapt them to changed environments is difficult and slow.
Why it matters
Adaptive manipulation learning reduces costly manual programming for industrial and service robots.
What the AI did
Two model-free reinforcement learning algorithms learned a grasping task from simulation data, with a fine-tuning procedure for continuous adaptation to new environments.
Results
Learned policy demonstrated generalizable across multiple object geometries and initial configurations; tested on a real Franka Emika Panda robot (abstract truncated before final outcome).
Impact
Evidence supports sim-to-real transfer of an RL grasping policy on a physical robot.
Record
| Venue | Autonomous Robots |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Asad Ali Shahid, Dario Piga, Francesco Braghin, Loris Roveda |
| Citations | 66 (OpenAlex, at import time) |
| References | 59 |
| Model or method | Proximal Policy Optimization (PPO) and Soft Actor-Critic (SAC) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1007/s10514-022-10034-z |
| Open access | hybrid |
| Keywords | reinforcement learning, robotic manipulation, PPO, SAC, sim-to-real, grasping |
Importance rating
real but narrower advance
Sim-to-real RL grasping shown on one robot/task; reduces manual robot programming, but abstract gives no scale of affected robots, industry, or adoption.
More in Robotics & Autonomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Autonomous robotic laparoscopic surgery for intestinal anastomosis | 2022 | 384 |
7
|
| 2 | Solving Rubik's Cube with a Robot Hand | 2019 | 569 |
6
|
| 3 | PaLM-E: An Embodied Multimodal Language Model | 2023 | 346 |
6
|
| 4 | Assistive Navigation Using Deep Reinforcement Learning Guiding Robot With UWB/Voice Beacons and Semantic Feedbacks for Blind and Visually Impaired People | 2021 | 48 |
6
|
| 5 | Towards Deep Radar Perception for Autonomous Driving: Datasets, Methods, and Challenges | 2022 | 154 |
6
|