An official website of the United States government
A common failure mode for policies trained with imitation is compounding execution errors at test time. When the learned policy encounters states that are not present in the expert demonstrations, the policy fails, leading to degenerate behavior. The Dataset Aggregation, or DAgger approach to this problem simply collects more data to cover these failure states. However, in practice, this is often prohibitively expensive. In this work, we propose Diffusion Meets DAgger (DMD), a method that reaps the benefits of DAgger but without the cost, for eye-in-hand imitation learning problems. Instead of collecting new samples to cover out-of-distribution states, DMD uses recent advances in diffusion models to synthesize these samples. This leads to robust performance from few demonstrations. We compare DMD against behavior cloning baseline across four tasks: pushing, stacking, pouring, and hanging a shirt. In pushing, DMD achieves 80% success rate with as few as 8 expert demonstrations, where naive behavior cloning reaches only 20%. In stacking, DMD succeeds on average 92% of the time across 5 cups, versus 40% for BC. When pouring coffee beans, DMD transfers to another cup successfully 80% of the time. Finally, DMD attains 90% success rate for hanging shirt on a clothing rack.
more »
« less
DART: Noise Injection for Robust Imitation Learning
One approach to Imitation Learning is Behavior Cloning, in which a robot observes a supervisor and infers a control policy. A known problem with this “off-policy” approach is that the robot’s errors compound when drifting away from the supervisor’s demonstrations. On-policy, techniques alleviate this by iteratively collecting corrective actions for the current robot policy. However, these techniques can be tedious for human supervisors, add significant computation burden, and may visit dangerous states during training. We propose an off-policy approach that injects noise into the supervisor’s policy while demonstrating. This forces the supervisor to demonstrate how to recover from errors. We propose a new algorithm, DART (Disturbances for Augmenting Robot Trajectories), that collects demonstrations with injected noise, and optimizes the noise level to approximate the error of the robot’s trained policy during data collection. We compare DART with DAgger and Behavior Cloning in two domains: in simulation with an algorithmic supervisor on the MuJoCo tasks (Walker, Humanoid, Hopper, Half-Cheetah) and in physical experiments with human supervisors training a Toyota HSR robot to perform grasping in clutter. For high dimensional tasks like Humanoid, DART can be up to 3x faster in computation time and only decreases the supervisor’s cumulative reward by 5% during training, whereas DAgger executes policies that have 80% less cumulative reward than the supervisor. On the grasping in clutter task, DART obtains on average a 62% performance increase over Behavior Cloning.
more »
« less
- Award ID(s):
- 1734633
- PAR ID:
- 10063838
- Date Published:
- Journal Name:
- CoRL
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Most existing policy learning solutions require the learning agents to receive high-quality supervision signals, e.g., rewards in reinforcement learning (RL) or high-quality expert demonstrations in behavioral cloning (BC). These quality supervisions are either infeasible or prohibitively expensive to obtain in practice. We aim for a unified framework that leverages the available cheap weak supervisions to perform policy learning efficiently. To handle this problem, we treat the weak supervision'' as imperfect information coming from a peer agent, and evaluate the learning agent's policy based on a correlated agreement'' with the peer agent's policy (instead of simple agreements). Our approach explicitly punishes a policy for overfitting to the weak supervision. In addition to theoretical guarantees, extensive evaluations on tasks including RL with noisy reward, BC with weak demonstrations, and standard policy co-training (RL + BC) show that our method leads to substantial performance improvements, especially when the complexity or the noise of the learning environments is high.more » « less
-
Robot Imitation Learning (IL) is a crucial technique in robot learning, where agents learn by mimicking human demonstrations. However, IL encounters scalability challenges stemming from both non-user-friendly demonstration collection methods and the extensive time required to amass a sufficient number of demonstrations for effective training. In response, we introduce the Augmented Reality for Collection and generAtion of DEmonstrations (ARCADE) framework, designed to scale up demonstration collection for robot manipulation tasks. Our framework combines two key capabilities: 1) it leverages AR to make demonstration collection as simple as users performing daily tasks using their hands, and 2) it enables the automatic generation of additional synthetic demonstrations from a single human-derived demonstration, significantly reducing user effort and time. We assess ARCADE's performance on a real Fetch robot across three robotics tasks: 3-Waypoints-Reach, Push, and Pick-And-Place. Using our framework, we were able to rapidly train a policy using vanilla Behavioral Cloning (BC), a classic IL algorithm, which excelled across these three tasks. We also deploy ARCADE on a real household task, Pouring-Water, achieving an 80% success rate.more » « less
-
Learning from Demonstration (LfD) is a promising approach to enable Multi-Robot Systems (MRS) to acquire complex skills and behaviors. However, the intricate interactions and coordination challenges in MRS pose significant hurdles for effective LfD. In this paper, we present a novel LfD framework specifically designed for MRS, which leverages visual demonstrations to capture and learn from robot-robot and robot-object interactions. Our framework introduces the concept of Interaction Keypoints (IKs) to transform the visual demonstrations into a representation that facilitates the inference of various skills necessary for the task. The robots then execute the task using sensorimotor actions and reinforcement learning (RL) policies when required. A key feature of our approach is the ability to handle unseen contact-based skills that emerge during the demonstration. In such cases, RL is employed to learn the skill using a classifier-based reward function, eliminating the need for manual reward engineering and ensuring adaptability to environmental changes. We evaluate our framework across a range of mobile robot tasks, covering both behavior-based and contact-based domains. The results demonstrate the effectiveness of our approach in enabling robots to learn complex multi-robot tasks and behaviors from visual demonstrations.more » « less
-
Using the context of human-supervised object collection tasks, we explore policies for a robot to seek assistance from a human supervisor and avoid loss of human trust in the robot. We consider a human-robot interaction scenario in which a mobile manipulator chooses to collect objects either autonomously or through human assistance; while the human supervisor monitors the robot’s operation, assists when asked, or intervenes if the human perceives that the robot may not accomplish its goal. We design an optimal assistance-seeking policy for the robot using a Partially Observable Markov Decision Process (POMDP) setting in which human trust is a hidden state and the objective is to maximize collaborative performance. We conduct two sets of human-robot interaction experiments. The data from the first set of experiments is used to estimate POMDP parameters, which are used to compute an optimal assistance-seeking policy that is used in the second experiment. For most participants, the estimated POMDP reveals that humans are more likely to intervene when their trust is low and the robot is performing a high-complexity task; and that the robot asking for assistance in high-complexity tasks can increase human trust in the robot. Our experimental results show that the proposed trust-aware policy yields superior performance compared with an optimal trust-agnostic policy.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
