Platform to Test and Evaluate Human-in-the-Loop Telemanipulation Schemes for Autonomous Unmanned Aerial Systems

Abstract

In recent years, remote multi-agent human-robot communication and remote agent telemanipulation have thrived considerably in line with the rise of autonomy since Tesla‘s invention of the "distant operator" with a patent (US613809A) in 1898 for controlling wireless drones remotely. As a result of taking on more and more tasks that require decision-making skills, vehicles are becoming more and more automated using intelligent control systems that are getting better with advanced sensors and actuators using AI. While the ultimate goal of these systems is full autonomy, it may take years/decades to make these systems thoroughly integrated into mixed traffic due to the varying expectations of stakeholders and their level of trust in autonomous vehicles (AVs). To expedite this integration and consequently exploit the benefits of these automated systems, in all future scenarios of fully autonomous ground or air vehicle networks, human intervention is expected to take place in the form of "human-in-the-loop (HITL)" remote immediate involvement. In this direction, a scalable agent-based platform was designed for the telemanipulation of autonomous unmanned aerial vehicles (A-UAVs) from the perspective of human-multi-robot architecture by which various HITL telemanipulation schemes can be tested, evaluated and improved. This report explains the design and infrastructure of this platform with its fundamental components.