Cooperative Trajectory Planning: Principles for Human-Machine System Design on Trajectory Level

TL;DR

This paper presents an approach of finding a joint trajectory by modelling cooperative trajectory planning as an agreement process, and proposes concepts to implement cooperative trajectory planning as an agreement process.

Abstract

This paper explores cooperative trajectory planning approaches within the context of human-machine shared control. In shared control research, it is typically assumed that the human and the automation use the same reference trajectory to stabilize the coupled system. However, this assumption is often incorrect, as they usually follow different trajectories, causing control conflicts at the action level that have not been widely researched. To address this, it is logical to extend shared control concepts to include human-machine interaction at the trajectory-level before action execution, resulting in a unified reference trajectory for both human and automation. This paper begins with a literature overview on approaches of cooperative trajectory planning. It then presents an approach of finding a joint trajectory by modelling cooperative trajectory planning as an agreement process. A generally valid system structure is proposed for this purpose. Finally, it proposes concepts to implement cooperative trajectory planning as an agreement process.

Figures (3)

• Figure 1: System structure of shared control without a common reference trajectory (left), with a common, given reference trajectory (middle) and with an additional trajectory level to find a common reference trajectory jointly (right).
• Figure 2: Comparison of five different instantiations of fusions of trajectory requests.
• Figure 3: System structure for cooperative trajectory planning at the trajectory level.