DIU: Collaborative Autonomy for Networked Aircraft

Suspense Date: 4 June 2021 Description: Collaborative Autonomy for Networked Aircraft The Department of Defense (DoD) seeks commercial solutions to develop artificial intelligence (AI) algorithms for networked collaborative autonomous (NCA) air platforms in a digital engineering environment.

Category

Opportunity

DoD Communities of Interest

Artificial Intelligence

Subject

Collaborative Autonomy for Networked Aircraft

Due Date

2021-06-06 23:59:59 US/Eastern Time

Government Organization

DIU

Description
Collaborative Autonomy for Networked Aircraft
Responses Due By
2021-06-06 23:59:59 US/Eastern Time

The Department of Defense (DoD) seeks commercial solutions to develop artificial intelligence (AI) algorithms for networked collaborative autonomous (NCA) air platforms in a digital engineering environment. Solutions must be capable of coordinating and prioritizing actions in complex and contested environments. Submitters should clearly explain how their proposed algorithms and approach are consistent with the DoD Ethical AI Principles and DoD 3000.09 guidelines. While these algorithms extend to a variety of use cases, this specific prototype evaluation will be focused on coordinating long-range, high-speed, fixed wing aerial platforms operating in contested environments.

The program will be structured as a series of prototype events where selected performers will deploy their algorithms in an unclassified, live-virtual-constructive development environment based upon the Advanced Framework for Simulation, Integration, and Modeling (AFSIM). The environment will include a clearly defined state and action space, and scenarios will include mission objectives, potential threats, and evaluation criteria. The program structure will include approximately three intermediate events and a final prototype competition. Performers’ algorithms will compete against both each other and increasingly sophisticated government reference algorithms in a variety of scenarios to evaluate their effectiveness.

Solutions will provide autonomous path planning in three dimensions -- to include collaborative formations for increasing survivability and threat avoidance as well as dynamic task allocation in an objective area. Input data (objective and threat locations, environmental uncertainties, etc) will be provided in a structured manner before and during mission execution, and a vehicle control interface will be defined by the Government. Therefore, this solicitation is not seeking computer vision, perception algorithms, or control algorithms, but is focused on collaborative autonomy behaviors that adapt to changes in the world state in order to achieve increased survivability and effectiveness. Examples include path or route planning, formation control and healing, and synchronized arrival.

Performers will be provided a baseline Government Reference Architecture with mapping for specific functional modules to be described in Phase Two of this solicitation. Performers can select to modify one, multiple, or all modules, or to modify the architecture as a whole. Preference will be given to vendors, or teams of vendors, that show a capability for generating an end-to-end solution. Vendors who propose solutions using a teaming and/or prime subcontractor arrangement, which combine multiple solutions into one, are highly encouraged to submit such solutions individually, as well as together.

Initial Task Components and Guidelines:

Performers will be provided an initial version of the Government Reference Architecture and Interface Standards, but are encouraged to make modifications to increase mission effectiveness.
Performers will deliver their modular algorithms or complete agent to the Government team for evaluation in the simulation environment.
Algorithms and frameworks will be portable to a variety of platforms.
Performers will interact with government-provided operators throughout the period of performance to collaborate on solution design and demonstrate the incorporation of debrief data and feedback.
Where possible, use open standards for data ingest, storage, and exchange, as well as for the training, deployment, and transfer of AI algorithms.
Vendors will require the ability to handle Controlled Unclassified Information (CUI)/Distribution D material during prototype phase.

Website

https://www.diu.mil/work-with-us/open-solicitations