Agenda

Welcome & Opening Remarks (Day 1 - Main Hall)

Bogdan Epureanu
Director, Automotive Research Center

David Gorsich
Chief Scientist, U.S. Army DEVCOM Ground Vehicle Systems Center (GVSC)

The Honorable Gary Peters (Day 1 - Main Hall)
United States Senate

Dr. Alec D. Gallimore (Day 1 - Main Hall)
Robert J. Vlasic Dean of Engineering at the University of Michigan

Keynote: (Day 1 - Main Hall)

Michael Cadieux
Director, U.S. Army DEVCOM GVSC

Keynote: (Day 1 - Main Hall)

Christopher Davey
Global Manager of Systems Engineering, System Safety, Modelling & Simulation at Ford Motor Company

Break

Case Study 1

There is no "I" in Team: Optimal task allocation in human-autonomy teaming

Abstract: Autonomous vehicles are increasingly thought of as team members alongside humans in both military and civilian applications. Such autonomous agents are capable of handling dangerous tasks but are limited in their reactions to unforeseen events. At the same time, humans have more adaptive and creative problem-solving skills but are limited in terms of handling some specific tasks and managing cognitive loads. The inclusion of autonomy within a team requires a significant effort to train the agents and dynamically distribute tasks among the agents to perform optimally during operations.

In this case study, we brought together three projects and constructed a unique framework to train a team of heterogeneous agents, composed of both humans and autonomous agents, to reliably perform tasks in uncertain environments. A computational trust model for multi-agent teams was created and deployed in trust-based path planning algorithms. The cost and limitations of the mobility of the agents was accounted for when training the team in a synthetic environment. An artificial intelligence algorithm was then developed for autonomous agents to learn how to collaborate with humans and other autonomous agents through reinforcement learning. To showcase the application of the developed algorithms, a disaster relief scenario was simulated in a high-fidelity game engine environment where a human interacts with the environment in real-time using virtual reality. An adaptive algorithm was developed to assist the humans in their decision-making and improve their performance by continuously evaluating the human’s cognitive task loads. The heterogeneity of the team was described by the differences in agent capabilities of task handling, sensing, and communication, as well as the level of risk aversion in humans’ decision-making processes. Results of this study show that the trained autonomous agents using the developed algorithms can reliably collaborate with humans and clear all the assigned tasks in a complex environment. In this study, the team performance when the human-autonomy communication used an adaptive interface based on a data-driven cognitive task load model increased by over 180% compared to a heuristic interface, and over 72% compared to a fixed and most detailed interface. The proposed framework is a basis for further developments and design of human-autonomy teams.

Break

Technical Session 1

Session 1.A

Session Chairs:
Jon Smereka, Mike Cole

Session 1.B

Session Chairs:
Matt Castanier, Steve Rapp

Introductions (Day 2 - Main Hall)

Kira Barton & Ram Vasudevan
Associate Directors, Automotive Research Center

Keynote: (Day 2 - Main Hall)

Dr. Patrick Baker
Director, U.S. Army DEVCOM Army Research Laboratory

Keynote: (Day 2 - Main Hall)

Tony Bromwell
Vice President, Engineering Operations at MSC Software

Keynote: (Day 2 - Main Hall)

Kevin Dutcher
Sr Systems Engineer, Ricardo Defense

Case Study 2

Right Resources, Right Place, Right Time: Using real-time physics-based simulation to optimize high-level mission planning under uncertainty

Abstract: As Army vehicles become more intelligent with autonomous behaviors, planning their missions under uncertainties becomes more complex. On the one-hand, low-level `path planning algorithms’ solve the challenge of finding an optimal trajectory between a pair of source and destination while considering obstacles, and provide closed-loop control signals to each vehicle so that they can follow the trajectory with minimum deviations. On the other hand, a high-level ‘mission planning algorithm’ provides the sequence of visits for the vehicle while considering points of interests, environment uncertainties, and other operational constraints. Though these low-level and high-level algorithms work in tandem, due to overwhelming complexities, the research works have traditionally been done independent to each other. In this case study, we address this gap by bringing two ARC projects together to integrate real-time physics-based simulation of autonomous ground vehicles with high-level mission planning algorithms.

The summary of the case study are as follows: best and worst scenarios inferred from data summarization may not necessarily be the best and worst scenarios for mission planning; for a fixed importance of distance and HRI costs, the mission plan across scenarios from low-level planning may change, and this is specifically observed in scenarios with lower goal speed. The deviations in mission plan due to scenarios is around 25%; and also, deviations within scenarios due to weights for distance and HRI costs is at 35%. These imply the importance of synergy between low-level and high-level planning algorithms.

Technical Session 2 - Part I

Session 2.A

Session Chairs:
Victor Paul, Matt Castanier

Session 2.B

Session Chairs:
Katie Sebek, David Lamb

Technical Session 2 - Part II

Session 2.C

Session Chairs:
Matt Castanier, Jon Smereka

Session 2.D

Session Chairs:
Vamshi Korivi, Steve Rapp

Bogdan Epureanu
Director, Automotive Research Center

David Gorsich
Chief Scientist, U.S. Army DEVCOM Ground Vehicle Systems Center (GVSC)