DoD Laboratory Autonomy Research Pilot Initiative: Seven Proposals Selected
Development and deployment of autonomous technology is a priority across the Department of Defense. In November 2012, to complement our contracted research in autonomy with industry and academia, the department established the Autonomy Research Pilot Initiative (ARPI). This initiative is an experimental effort to attract internal DoD ideas from laboratory bench-level scientists and engineers, and encourage cross-service communication and collaboration on autonomy-related research topics.
Guided by feedback from operational experience and evolving mission requirements, seven proposals were selected out of 54 submissions to be a part of this $45 million multi-year activity. These proposals feature either unique solutions for the cross-domain applicability of the products or demonstrate new approaches to more established problems. The seven proposals selected are:
- Exploiting Priming Effects in Autonomous Cognitive Systems: Develops machine perception that is relatable to the way a human perceives an environment (Navy Center for Applied Research in Artificial Intelligence; Army Research Laboratory)
- Autonomous Squad Member: Integrates machine semantic understanding, reasoning and understanding, perception into a ground robotic system (Army Research Laboratory; Naval Research Laboratory, Navy Center for Applied Research in Artificial Intelligence)
- Autonomy for Adaptive Collaborative Sensing: Develops intelligent ISR capability for sensing platforms to have capability to find/track targets (Air Force Research Laboratory; Army Research Laboratory; Naval Research Laboratory)
- Realizing Autonomy via Intelligent Adaptive Hybrid Control: Develops flexible UAV operator interface enabling operator to "call a play" or manually control system (Air Force Research Laboratory; SPAWAR, Naval Research Laboratory; Army Research Laboratory)
- Autonomy for Air Combat Missions: Mixed Human/UAV Teams: Develops goal-directed reasoning, machine learning and operator interaction techniques to enable management of multiple, team UAVs (Air Force Research Laboratory; Naval Research Laboratory; Naval Air Warfare Center; Army Research Laboratory)
- A Privileged Sensing Network-Revolutionizing Human-Autonomy Integration: Develops integrated human sensing capability to enable the human-machine team (Army Research Laboratory, Army Tank Automotive Research Center; Air Force Research Laboratory)
- Autonomous Collective Defeat of Hard and Deeply Buried Targets: Develops small UAV teaming algorithms to enable systems to autonomously search a cave (Air Force Research Laboratory; Army Research Laboratory; Defense Threat Reduction Agency)
The goal of ARPI is to advance technologies that will result in autonomous systems that provide more capability to warfighters, lessen the cognitive load on operator/supervisors, and lower overall operational cost. The potential cross-cutting advances of this initiative in multiple domains provide an exciting prospect for interoperability among the military services as well as potentially meeting future acquisitions requirements.
Invitation for Proposal and Four Technology Development Efforts
The Invitation for Proposals (IFP) was aimed at funding technology development efforts that address one or more of the four
Priority Steering Council (PSC)-identified technical challenges.
1. Human/Autonomous Systems Interaction and Collaboration, including but not limited to:
- Robust Cognitive and Neurological Models that can model human interaction and teaming with autonomous systems
- Integration of Autonomy, Artificial Intelligence and Human Cognitive, or Other Human Models
- Optimized Trust in Automation/Transparency
- Principled Control Station Human Factors Engineering
- Advanced Feedback Interfaces to Maximize Common Perception between Human(s) and Agent(s)
- Secure Communication between Human(s) and Agent(s) and understanding of intent and actions of human team members, adversaries and bystanders.
- Advanced Control System Interfaces
2. Scalable Teaming of Multiple Autonomous Systems, including but not limited to:
- Shared Problem Solving/Reasoning between Agents
- Shared Perception between Agents
- System Health Management/Attrition Management
- Secure Communication between Multiple Agents
- Scalable collaboration among heterogeneous teams.
3. Machine Reasoning, Perception, and Intelligence, including but not limited to:
- Data-Driven Analytics
- Sensor/Data Decision Models
- Advanced Algorithms to Enable Robust Operations in unstructured environments including machine learning
- Contingency-based Control Strategies
- Adaptive Guidance and Control integration with higher level reasoning, decision-making, learning.
- Domain Management and Mission Control
- Integrated Contextual Decision Making
- Cognitive, intelligent, and adaptive computing paradigm/platform
4. Test, Evaluation, Validation, and Verification, including but not limited to:
- Simulation Test Beds for Human-Agent Teaming
- Live Test Beds for Human-Agent Teaming
- Simulation Test Beds for Operation in Complex, Contested Environments
- Live Test Beds for Operation in Complex, Contested Environments
- Simulation Test Beds for Controlled, Coordinated Actions by Multiple Agents
- Live Test Beds for Controlled, Coordinated Actions by Multiple Agents
- Methods/Strategy to Test and Evaluate Autonomous Systems/Subsystems
- Methods/Strategy to Test and Evaluate Human-Agent Interfaces
- Methods/Strategy for Validation and Verification of Computer Models/Logic, Algorithms, and Integrated Software Tools
Who was Invited to Participate
All DoD laboratories, with the exception of medical laboratories, were eligible to submit proposals. Laboratories included technical directorates, engineering centers, and warfare centers. Other agencies and organizations were allowed to participate as partners, but the “intellectual” work was largely done within the DOD labs. The lead organization submitting the proposal was a DOD laboratory.
About this Effort
The ARPI is a pilot test of an Office of the Secretary of Defense sponsored innovation program, directed by the Assistant Secretary of Defense for Research and Engineering, and executed by the Services, with support from the DoD Priority Steering Council (PSC) for Autonomy and the Research and Engineering Executive Committee.