XPONENTIAL 2023
SPEAKER SCHEDULE
A workshop led by:
Ross Eaton
Director of Marine Systems
and Principal Scientist
James Everson
Marine Operations
Supervisor and Engineer
Jacob Riedel
Software Engineer in Sensing, Perception,
and Applied Robotics
The Design Dream, the Operational Nightmare: How to Ensure that Autonomy Designers’ Efforts Deliver on Operators’ Dreams
Tuesday, May 9, 2023: 11:00–12:30 PM
Room: 110/112
Session Number: XPO23-WK05
Autonomy offers huge potential benefits across society, in the air, land, sea, and space domains. Everyone involved in autonomy development, from designers to end users, regulators, and those involved in deploying and maintaining the supporting infrastructure, is doing their best to realize those benefits. So why is autonomy slow to live up to its promise, despite everyone’s best intentions? We suggest that one continuing challenge is communication failures that lead to mismatches between the understanding and assumptions of stakeholders involved in different stages of autonomy design, development, and the operationalization process.
This workshop will take a playful look at places these communication breakdowns can occur through a fun, hands-on design and test process. Participants will discuss real-world examples of autonomy successes and failures and look for common through lines that lead to these outcomes. A set of key lessons for increasing operational success will be shared, and the workshop will conclude with a discussion of how to apply these lessons to the development and operationalization of adaptive autonomy solutions across the expanding range of operating domains.
Presented by:
David Koelle
Director of Engineering and
Principal Software Engineer
A Staged Blueprint for Realizing
Collaborative Autonomy
A Staged Blueprint for Realizing Collaborative Autonomy
Tuesday, May 9, 2023: 2:00–2:25 PM
Room: 104/106
Session Number: XPO23-4848
UAS autonomy enables vehicles to make their own decisions about navigation and other basic maneuvers, thus removing the need for operators to remotely micromanage (via joystick) basic performance of these systems. Collaborative autonomy explores the use of multiple UAS working as a team to address mission-level goals and objectives in a way that optimizes the use of each vehicle’s capabilities. Freed from management of low-level control, a single human operator assigned to the team is able to focus on assessment of higher-order results and related tasking in real-time. For example, a heterogeneous group of UAS optimized for agriculture inspection could be comprised of a high-flying UAS that can scan fields for distressed and diseased crops, other UAS outfitted with effectors to apply specific remediations to problem plants, and even ground vehicles that can be called on for additional assistance to increase the speed, precision, and cost of applying fertilizer, pesticides, weed removal, and other needs. In this talk, we will discuss a blueprint for meeting the challenges ahead of us in the wide-scale adoption of collaborative autonomy, including a series of suggested interim stages that offer increasing value (while limiting risk) to the interested community of operators, developers, and regulators.
Presented by:
Dr. Arjuna Balasuriya
Marine Robotics
Senior Scientist
Challenges and Progress in
Behavior-Based Adaptive Autonomy
Tuesday, May 9, 2023: 2:30–2:55 PM
Room: 110/112
Session Number: XPO23-4833
Today’s uncrewed platforms are typically operated by humans using remote control to guide every detailed aspect of a mission. However, as missions become more complex, there are many scenarios (particularly in the marine and ground domains) in which operators are unable to communicate with these uncrewed platforms in real-time (due to adverse environmental conditions, regulatory restrictions on communications in ecologically sensitive areas, active interference by adversaries, or the desire to remain covert), making it challenging to re-calibrate and update mission and control parameters on-the-fly. Fortunately, significant technical advances in onboard computing power and enhanced sensors offer a pathway to a level of autonomy that can overcome such communications limitations. Behavior-based autonomy architectures enable onboard mission autonomy software to select optimal behaviors for the mission at hand, assess the evolving state of the mission, and adapt to changing environmental conditions. Developers and operators of autonomy systems for challenging environments will gain insights into the importance of onboard behavior and parameter adaptation to the success of long-duration uncrewed platform missions in unpredictable situations and unknown environments. Designers will benefit from a discussion of how multi-objective optimization and reinforcement learning techniques are currently being used to enable assured autonomy onboard these uncrewed platforms.
A panel led by:
Rich Wronski
Division Vice President
and Principal Scientist
Daniel W. Stouch
Principal Scientist and Director of Space and Airborne Systems
Stephanie Kane
Principal Scientist and Director of UX Innovations
Ross Eaton
Director of Marine Systems
and Principal Scientist
Dr. Madison Clark-Turner
Scientist in Sensing, Perception, and
Applied Robotics
Achieving the Blueprint for Autonomy: The Way Forward from a Trusted Source
Tuesday, May 9, 2023: 4:00–4:45 PM
Room: 203
Session Number: XPO23-EG05
There exist myriad challenges to overcome in ultimately achieving the Blueprint for Autonomy. Some are complex technically—the opaqueness of AI/ML, the multidimensionality of adaptive autonomy (across land, sea, air, and space domains), and of course verification and validation of these continuously evolving systems. Human engagement with these platforms remains (or should remain!) central to any discussion of autonomy, driving much deep thinking these days by the cognitive engineering community engaged in development of the systems and supporting workflows. Critical to adoption is the need to establish trust in this evolving autonomy—a tall order when there are so many unknowns, not only to end users, operators, and policy makers, but even to developers themselves.
Join us for a practical discussion delving into these very issues. Our panel of domain experts will open that black box, and you’ll take away answers to some of your most pressing questions (and maybe even a clear pathway to solving your own autonomy challenges) no matter your role in the autonomy ecosystem.
At XPONENTIAL 2023, we are featuring our leading-edge adaptive and collaborative autonomy products in the maritime, ground, air, and space domains.
