REVIEW 7 cited by
Continuously Improving Mobile Manipulation with Autonomous Real-World RL
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Continuously Improving Mobile Manipulation with Autonomous Real-World RL
read the original abstract
We present a fully autonomous real-world RL framework for mobile manipulation that can learn policies without extensive instrumentation or human supervision. This is enabled by 1) task-relevant autonomy, which guides exploration towards object interactions and prevents stagnation near goal states, 2) efficient policy learning by leveraging basic task knowledge in behavior priors, and 3) formulating generic rewards that combine human-interpretable semantic information with low-level, fine-grained observations. We demonstrate that our approach allows Spot robots to continually improve their performance on a set of four challenging mobile manipulation tasks, obtaining an average success rate of 80% across tasks, a 3-4 improvement over existing approaches. Videos can be found at https://continual-mobile-manip.github.io/
Forward citations
Cited by 7 Pith papers
-
SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning
A video-language model with per-timestep spatiotemporal CoT and dense progress prediction can serve as the sole reward for zero-shot online robot RL on 24 unseen manipulation tasks.
-
FT-WBC: Learning Fault-Tolerant Whole-Body Control for Legged Loco-Manipulation
FT-WBC introduces a decoupled policy architecture with a Fault Estimator and Posture Adaptation Module that converts unstable arm-driven posture requests into safe base commands under actuator failures in legged manipulators.
-
Learning Process Rewards via Success Visitation Matching for Efficient RL
Success Visitation Matching uses a discriminator to turn sparse outcome rewards into dense process rewards by matching visitations of successful episodes, provably preserving the optimal policy and speeding up robotic...
-
From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning
Residual off-policy RL with selective BC regularization and value-guided sampling contracts a pretrained generative robot policy around successful actions, reaching high success on hard long-horizon tasks from pixels ...
-
TwinRL: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation
TwinRL expands RL exploration via digital twin reconstruction and twin RL warm-up to guide real-world learning, reaching near-100% success with 20 minutes of on-robot time across four tasks.
-
UMI-on-Air: Embodiment-Aware Guidance for Embodiment-Agnostic Visuomotor Policies
Embodiment-Aware Diffusion Policy steers a UMI-trained diffusion policy with controller tracking-cost gradients at inference time, improving aerial manipulation success in simulation and real flights.
-
FT-WBC: Learning Fault-Tolerant Whole-Body Control for Legged Loco-Manipulation
FT-WBC is a decoupled-policy framework that uses fault estimation and posture adaptation to synthesize compensatory gaits and preserve arm workspace in legged manipulators under actuator failures.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.