Learning Personalized Human-Aware Robot Navigation Using Virtual Reality Demonstrations from a User Study

Abstract

Foresighted robot navigation in dynamic indoor environments with cost-efficient hardware necessitates the use of a lightweight yet dependable controller. So inferring the scene dynamics from sensor readings without explicit object tracking is a pivotal aspect of foresighted navigation among pedestrians. In this paper, we introduce a spatiotemporal attention pipeline for enhanced navigation based on 2D lidar sensor readings. This pipeline is complemented by a novel lidar-state representation that emphasizes dynamic obstacles over static ones. Subsequently, the attention mechanism enables selective scene perception across both space and time, resulting in improved overall navigation performance within dynamic scenarios. We thoroughly evaluated the approach in different scenarios and simulators, finding good generalization to unseen environments. The results demonstrate outstanding performance compared to state-of-the-art methods, thereby enabling the seamless deployment of the learned controller on a real robot.

Learning Depth Vision-Based Personalized Robot Navigation From Dynamic Demonstrations in Virtual Reality

“…Fusing the potential of user demonstrations with a learning architecture lead to promising results in the field of robotic manipulation tasks [11]. Therefore, this is a key concept for our learning architecture and has proven success the field of robot navigation [2].…”

“…While in our previous work [2], we presented one of the first approaches at the intersection of both navigation and robot personalization, we now enhance the system by allowing to demonstrate navigation trajectories under dynamic motions and using only depth vision as input.…”

“…Where humans share the same environment with a mobile robot, the robot's navigation behavior significantly influences the comfort of interaction [2]. Consequently, basic obstacle avoidance approaches are insufficient to address individual preferences such as proxemics, trajectory shape, or area of navigation in a given environment, while being a key component to a successful navigation without question.…”

“…Recent advances in user-focused robot navigation have paved the way to personalized navigation controllers [2]. More specifically, a virtual reality interface paired with a reinforcement learning framework enables the demonstration and training of highly customizable navigation behaviors.…”

Handling Sparse Rewards in Reinforcement Learning Using Model Predictive Control

“…training process in case of sparse rewards [6]- [8], and human demonstrations specifically are commonly used in the literature, providing these demonstrations can be quite costly. In addition to that, human demonstrations typically require hardware equipment or virtual reality sets to provide the demonstrations [6], [9]- [11]. In this work, we therefore propose to use MPC as an experience source for RL in the case of sparse rewards.…”