Get Started
Case Studies

Goal Reasoning for Satellite Constellation Decision Making

Goal Reasoning for Satellite Constellation Decision Making

US Army LDAC faced the challenge of managing a complex, evolving IT environment critical to Army logistics, and Knexus delivered integrated, secure, and efficient IT support that enhanced system reliability and agility while ensuring timely access to essential logistics data.

Goal Reasoning for Satellite Constellation Decision Making
PROBLEM

Insufficient personnel to control an increasing number of satellites while performing an increasing number of mission objectives in maritime recon scenarios.

LDAC must provide the Army community with vital logistics data necessary for the planning, conducting and sustainment of warfighting capability worldwide.

Knexus
Solution

Automation of Surveillance Assets.

  • Build tool to convert 2010 data into the 2020 Census format and produce a test dataset that can be used for rapid experimentation.
  • Built and managed a Jenkins system that integrates with all Disclosure Avoidance Census code to ensure smoother production and code iteration.
  • Leveraged AWS Cloud Infrastructure to create versioned testing tools that allowed for reproducibility of tests on any current or historical version of the DAS system.
  • Brought scientific research code to production ready standards for improved performance, legibility, and flexibility.

Utilize Goal Reasoning (GR) techniques to automate imaging tasks effectively for satellites in a constellation (group of satellites), while managing dynamically changing collection goals and priorities.

Utilize an AI model to automate constellation Observation, Orientation, Decisions, and Actions.

Demonstrate possible AI agent response to various simulated dynamic events including satellite availability/blackouts, weather conditions, changing mission objectives/priorities, and unexpected observations/behavior.

Customer Outcome

Customer Outcome

Goal Reasoning for Satellite Constellation Decision Making

GR proven it can be used to automate 100s of satellites and 1000s of image tasks in maritime reconnaissance scenarios through extensive testing.

GR can task a constellation of 50 satellites over an area the size of the South China Sea while managing hundreds of goals within seconds proven through simulation.

By automating satellites, constellation management now requires significantly less manpower.

Additional Case Studies

Deep Reinforcement Learning

Read More

AI-Powered Multi-Asset, Multi- Doman Operations Planning and Replanning Software

Read More

Minefield Planning Evaluation and Gaming Suite

Read More

Information Technology Services (LITeS) III

Read More
Get Started