Technical Report: Analysis of Intervention Modes in Human-In-The-Loop (HITL) Teleoperation With Autonomous Ground Vehicle Systems

Abstract

Fully autonomous systems are human-out-of-the-loop systems that single-handedly determine the right course of action when given an autonomous task. In all future visions of operating networks of fully autonomous self-driving ground or aerial vehicles, humans are expected to intervene with some kind of remote instantaneous intervention role and ``Human-on-the-Loop (HOTL)'' and ``Human-in-the-Loop (HOTL)'' telemonitoring and telemanipulation is expected to establish a desired level of trust in AVs while they are interacting with a highly dynamic urban or aerial environment. Many studies envision a future with fully autonomous self-driving vehicles (FA-SDVs) with increasing penetration levels in mixed traffic. However, effective management of FA-SDVs in real-world use cases under highly uncertain conditions has not been examined sufficiently in the literature. This report, by covering the teleoperation collaboration modes between two intelligent agents — human telesupervisors (HTSs) and FA-SDVs — aims to close this gap.