Human-robot cross-training: Computational formulation, modeling and evaluation of a human team training strategyThe MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. CitationNikolaidis, Stefanos, and Julie Shah. "Human-robot cross-training: Computational formulation, modeling and evaluation of a human team training strategy." In 2013 8th ACM/IEEE International Conference on Human-Robot Interaction (HRI), 33-40. Institute of Electrical and Electronics Engineers, 2013.

Perspective-taking is the ability to perceive or understand a situation or concept from another individual's point of view, and is crucial in daily human interactions. Enabling robots to perform perspective-taking remains an unsolved problem; existing approaches that use deterministic or handcrafted methods are unable to accurately account for uncertainty in partially-observable settings. This work proposes to address this limitation via a deep world model that enables a robot to perform both perception and conceptual perspective taking, i.e., the robot is able to infer what a human sees and believes. The key innovation is a decomposed multi-modal latent state space model able to generate and augment fictitious observations/emissions. Optimizing the ELBO that arises from this probabilistic graphical model enables the learning of uncertainty in latent space, which facilitates uncertainty estimation from high-dimensional observations. We tasked our model to predict human observations and beliefs on three partially-observable HRI tasks. Experiments show that our method significantly outperforms existing baselines and is able to infer visual observations available to other agent and their internal beliefs.

We present a framework for automatically learning human user models from joint-action demonstrations that enables a robot to compute a robust policy for a collaborative task with a human. First, the demonstrated action sequences are clustered into different human types using an unsupervised learning algorithm. A reward function is then learned for each type through the employment of an inverse reinforcement learning algorithm. The learned model is then incorporated into a mixed-observability Markov decision process (MOMDP) formulation, wherein the human type is a partially observable variable. With this framework, we can infer online the human type of a new user that was not included in the training set, and can compute a policy for the robot that will be aligned to the preference of this user. In a human subject experiment (n = 30), participants agreed more strongly that the robot anticipated their actions when working with a robot incorporating the proposed framework (p < 0.01), compared to manually annotating robot actions. In trials where participants faced difficulty annotating the robot actions to complete the task, the proposed framework significantly improved team efficiency (p < 0.01). The robot incorporating the framework was also found to be more responsive to human actions compared to policies computed using a hand-coded reward function by a domain expert (p < 0.01). These results indicate that learning human user models from joint-action demonstrations and encoding them in a MOMDP formalism can support effective teaming in human-robot collaborative tasks.

Abstract-Human-robot collaboration presents an opportunity to improve the efficiency of manufacturing and assembly processes, particularly for aerospace manufacturing where tight integration and variability in the build process make physical isolation of robotic-only work challenging. In this paper, we develop a robotic scheduling and control capability that adapts to the changing preferences of a human co-worker or supervisor while providing strong guarantees for synchronization and timing of activities. This innovation is realized through dynamic execution of a flexible optimal scheduling policy that accommodates temporal disturbance. We describe the Adaptive Preferences Algorithm that computes the flexible scheduling policy and show empirically that execution is fast, robust, and adaptable to changing preferences for workflow. We achieve satisfactory computation times, on the order of seconds for moderatelysized problems, and demonstrate the capability for human-robot teaming using a small industrial robot.