Robotic Control via Embodied Chain-of-Thought Reasoning
- URL: http://arxiv.org/abs/2407.08693v2
- Date: Fri, 12 Jul 2024 19:19:34 GMT
- Title: Robotic Control via Embodied Chain-of-Thought Reasoning
- Authors: Michał Zawalski, William Chen, Karl Pertsch, Oier Mees, Chelsea Finn, Sergey Levine,
- Abstract summary: Key limitation of learned robot control policies is their inability to generalize outside their training data.
Recent works on vision-language-action models (VLAs) have shown that the use of large, internet pre-trained vision-language models can substantially improve their robustness and generalization ability.
We introduce Embodied Chain-of-Thought Reasoning (ECoT) for VLAs, in which we train VLAs to perform multiple steps of reasoning about plans, sub-tasks, motions, and visually grounded features before predicting the robot action.
- Score: 86.6680905262442
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: A key limitation of learned robot control policies is their inability to generalize outside their training data. Recent works on vision-language-action models (VLAs) have shown that the use of large, internet pre-trained vision-language models as the backbone of learned robot policies can substantially improve their robustness and generalization ability. Yet, one of the most exciting capabilities of large vision-language models in other domains is their ability to reason iteratively through complex problems. Can that same capability be brought into robotics to allow policies to improve performance by reasoning about a given task before acting? Naive use of "chain-of-thought" (CoT) style prompting is significantly less effective with standard VLAs because of the relatively simple training examples that are available to them. Additionally, purely semantic reasoning about sub-tasks, as is common in regular CoT, is insufficient for robot policies that need to ground their reasoning in sensory observations and the robot state. To this end, we introduce Embodied Chain-of-Thought Reasoning (ECoT) for VLAs, in which we train VLAs to perform multiple steps of reasoning about plans, sub-tasks, motions, and visually grounded features like object bounding boxes and end effector positions, before predicting the robot action. We design a scalable pipeline for generating synthetic training data for ECoT on large robot datasets. We demonstrate, that ECoT increases the absolute success rate of OpenVLA, the current strongest open-source VLA policy, by 28% across challenging generalization tasks, without any additional robot training data. Additionally, ECoT makes it easier for humans to interpret a policy's failures and correct its behavior using natural language.
Related papers
- $π_0$: A Vision-Language-Action Flow Model for General Robot Control [77.32743739202543]
We propose a novel flow matching architecture built on top of a pre-trained vision-language model (VLM) to inherit Internet-scale semantic knowledge.
We evaluate our model in terms of its ability to perform tasks in zero shot after pre-training, follow language instructions from people, and its ability to acquire new skills via fine-tuning.
arXiv Detail & Related papers (2024-10-31T17:22:30Z) - Latent Action Pretraining from Videos [156.88613023078778]
We introduce Latent Action Pretraining for general Action models (LAPA)
LAPA is an unsupervised method for pretraining Vision-Language-Action (VLA) models without ground-truth robot action labels.
We propose a method to learn from internet-scale videos that do not have robot action labels.
arXiv Detail & Related papers (2024-10-15T16:28:09Z) - Grounding Robot Policies with Visuomotor Language Guidance [15.774237279917594]
We propose an agent-based framework for grounding robot policies to the current context.
The proposed framework is composed of a set of conversational agents designed for specific roles.
We demonstrate that our approach can effectively guide manipulation policies to achieve significantly higher success rates.
arXiv Detail & Related papers (2024-10-09T02:00:37Z) - Affordance-Guided Reinforcement Learning via Visual Prompting [51.361977466993345]
Keypoint-based Affordance Guidance for Improvements (KAGI) is a method leveraging rewards shaped by vision-language models (VLMs) for autonomous RL.
On real-world manipulation tasks specified by natural language descriptions, KAGI improves the sample efficiency of autonomous RL and enables successful task completion in 20K online fine-tuning steps.
arXiv Detail & Related papers (2024-07-14T21:41:29Z) - Nonprehensile Planar Manipulation through Reinforcement Learning with
Multimodal Categorical Exploration [8.343657309038285]
Reinforcement Learning is a powerful framework for developing such robot controllers.
We propose a multimodal exploration approach through categorical distributions, which enables us to train planar pushing RL policies.
We show that the learned policies are robust to external disturbances and observation noise, and scale to tasks with multiple pushers.
arXiv Detail & Related papers (2023-08-04T16:55:00Z) - Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from
Offline Data [101.43350024175157]
Self-supervised learning has the potential to decrease the amount of human annotation and engineering effort required to learn control strategies.
Our work builds on prior work showing that the reinforcement learning (RL) itself can be cast as a self-supervised problem.
We demonstrate that a self-supervised RL algorithm based on contrastive learning can solve real-world, image-based robotic manipulation tasks.
arXiv Detail & Related papers (2023-06-06T01:36:56Z) - Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement
Learning [54.636562516974884]
In imitation and reinforcement learning, the cost of human supervision limits the amount of data that robots can be trained on.
In this work, we propose MEDAL++, a novel design for self-improving robotic systems.
The robot autonomously practices the task by learning to both do and undo the task, simultaneously inferring the reward function from the demonstrations.
arXiv Detail & Related papers (2023-03-02T18:51:38Z) - SQUIRL: Robust and Efficient Learning from Video Demonstration of
Long-Horizon Robotic Manipulation Tasks [8.756012472587601]
Deep reinforcement learning (RL) can be used to learn complex manipulation tasks.
RL requires the robot to collect a large amount of real-world experience.
S SQUIRL performs a new but related long-horizon task robustly given only a single video demonstration.
arXiv Detail & Related papers (2020-03-10T20:26:26Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.