LaND: Learning to Navigate from Disengagements

Kahn, Gregory; Abbeel, Pieter; Levine, Sergey

doi:10.48550/arxiv.2010.04689

Cited by 2 publications

(2 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reduce the burden on the supervisor, several robot-gated interactive IL algorithms such as SafeDAgger [9], EnsembleDAgger [5], LazyDAgger [10], and ThriftyDAgger [8] have been proposed, in which the robot actively solicits human interventions when certain criteria are met. Interactive reinforcement learning (RL) [53,54,55,56] is another active area of research in which robots learn from both online human feedback and their own experience. However, these interactive learning algorithms are designed for and primarily studied in the single-robot, single-human setting.…”

Section: Single-robot Single-human Interactive Learningmentioning

confidence: 99%

Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervision

Hoque¹,

Chen²,

Sharma³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Commercial and industrial deployments of robot fleets often fall back on remote human teleoperators during execution when robots are at risk or unable to make task progress. With continual learning, interventions from the remote pool of humans can also be used to improve the robot fleet control policy over time. A central question is how to effectively allocate limited human attention to individual robots. Prior work addresses this in the single-robot, single-human setting. We formalize the Interactive Fleet Learning (IFL) setting, in which multiple robots interactively query and learn from multiple human supervisors. We present a fully implemented open-source IFL benchmark suite of GPU-accelerated Isaac Gym environments for the evaluation of IFL algorithms. We propose Fleet-DAgger, a family of IFL algorithms, and compare a novel Fleet-DAgger algorithm to 4 baselines in simulation. We also perform 1000 trials of a physical block-pushing experiment with 4 ABB YuMi robot arms. Experiments suggest that the allocation of humans to robots significantly affects robot fleet performance, and that our algorithm achieves up to 8.8× higher return on human effort than baselines. See https: //tinyurl.com/fleet-dagger for code, videos, and supplemental material.

show abstract

Section: Single-robot Single-human Interactive Learningmentioning

confidence: 99%

Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervision

Hoque¹,

Chen²,

Sharma³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kurenkov et al [26] and Xie et al [48] leverage interventions from suboptimal supervisors to accelerate policy learning, but assume that the supervisors are algorithmic and thus can be queried cheaply. Amir et al [2], Kahn et al [23], Kelly et al [24], Spencer et al [42], and Wang et al [47] instead consider learning from human supervisors and present learning algorithms which utilize the timing and nature of human interventions to update the learned policy. By giving the human control for multiple timesteps in a row, these algorithms show improvements over methods that only hand over control on a state-by-state basis [6].…”

Section: Background and Related Workmentioning

confidence: 99%

LazyDAgger: Reducing Context Switching in Interactive Imitation Learning

Hoque

Balakrishna

Putterman

et al. 2021

Preprint

View full text Add to dashboard Cite

Corrective interventions while a robot is learning to automate a task provide an intuitive method for a human supervisor to assist the robot and convey information about desired behavior. However, these interventions can impose significant burden on a human supervisor, as each intervention interrupts other work the human is doing, incurs latency with each context switch between supervisor and autonomous control, and requires time to perform. We present LazyDAgger, which extends the interactive imitation learning (IL) algorithm SafeDAgger to reduce context switches between supervisor and autonomous control. We find that LazyDAgger improves the performance and robustness of the learned policy during both learning and execution while limiting burden on the supervisor. Simulation experiments suggest that LazyDAgger can reduce context switches by an average of 60% over SafeDAgger on 3 continuous control tasks while maintaining state-of-the-art policy performance. In physical fabric manipulation experiments with an ABB YuMi robot, LazyDAgger reduces context switches by 60% while achieving a 60% higher success rate than SafeDAgger at execution time.

show abstract

LaND: Learning to Navigate from Disengagements

Cited by 2 publications

References 31 publications

Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervision

Fleet-DAgger: Interactive Robot Fleet Learning with Scalable Human Supervision

LazyDAgger: Reducing Context Switching in Interactive Imitation Learning

Contact Info

Product

Resources

About