Environment Agnostic Goal-Conditioning, A Study of Reward-Free Autonomous Learning

In this paper we study how transforming regular reinforcement learning environments into goal-conditioned environments can let agents learn to solve tasks autonomously and reward-free. We show that an agent can learn to solve tasks by selecting its own goals in an environment-agnostic way, at training times comparable to externally guided reinforcement learning. Our method is independent of the underlying off-policy learning algorithm. Since our method is environment-agnostic, the agent does not value any goals higher than others, leading to instability in performance for individual goals. However, in our experiments, we show that the average goal success rate improves and stabilizes. An agent trained with this method can be instructed to seek any observations made in the environment, enabling generic training of agents prior to specific use cases.

Impact of Gaze-Based Interaction and Augmentation on Human-Robot Collaboration in Critical Tasks

We present a user study analyzing head-gaze-based robot control and foveated visual augmentation in a simulated search-and-rescue task. Results show that foveated augmentation significantly improves task performance, reduces cognitive load by 38%, and shortens task time by over 60%. Head-gaze patterns analysed over both the entire task duration and shorter time segments show that near and far attention capture is essential to better understand user intention in critical scenarios. Our findings highlight the potential of foveation as an augmentation technique and the need to further study gaze measures to leverage them during critical tasks.