SBIR-STTR Award

Weapons Automated Artificial Intelligence Planner (WAAP)
Award last edited on: 9/10/2023

Sponsored Program
SBIR
Awarding Agency
DOD : AF
Total Award Amount
$1,500,365
Award Phase
2
Solicitation Topic Code
AF212-0008
Principal Investigator
Ben Baysinger

Company Information

GBL Systems Corporation

760 Paseo Camarillo Suite 401
Camarillo, CA 93010
   (805) 987-4345
   jimb@gblsys.com
   www.gblsys.com
Location: Single
Congr. District: 26
County: Ventura

Phase I

Contract Number: 2022
Start Date: ----    Completed: 4/8/2022
Phase I year
2022
Phase I Amount
$1
Direct to Phase II

Phase II

Contract Number: N/A
Start Date: 4/8/2024    Completed: 4/8/2022
Phase II year
2022
(last award dollars: 1694359770)
Phase II Amount
$1,500,364

Increasingly complex and sophisticated threat systems and complicated mission areas and planning parameters pose an ever-growing challenge for mission and weapon planners. Compounding the problem are legacy processes and software such as the Joint Mission Planning Software (JMPS), which cannot adequately process and incorporate the large quantity of data necessary for modern mission planning. GBL Systems proposes the Weapons Automated Artificial Intelligence Planner (WAAP), a layered technology approach that applies modern automation and machine learning (ML) techniques to this complex mission/weapons planning domain. WAAP scales gracefully with increasing mission complexity and seamlessly incorporates vast quantities of data while reducing planner cognitive burden, increasing weapons planning throughput, and providing robust extensibility. The WAAP design builds upon proven artificial intelligence (AI) and ML technology to improve Weapons Strike Planning for the Next-Generation Open Mission System (NOMS) architecture. WAAP will eschew computationally expensive exhaustive search, instead rapidly providing “optimal enough” solutions within reasonable planning timeframes to meet mission and weapons planning goals. Three critical components will be featured. First, a Synthetic Data Generation Harness can process and encapsulate “black box” legacy weapon flyout models. Second, a Learner that can analyze unstructured and structured data alike, featuring advanced automated machine learning. Third, a Replica that iteratively improves weapons planning using ML models and AI techniques. WAAP will address five key Improvement Areas for NOMS and weapons planning: Increasing process automation to eliminate manual weapons planning and decrease planning times while improving quality of results; Development of core weapons planning capabilities via a microservice architecture to result in a more robust and reliable planning experience that can quickly be updated; Integrating ML and AI for weapons delivery and threat avoidance with a continually enhanced planning workflow; Developing a next-generation flyout model service for common weapon models that enable the quick and easy extension for new weapons; And, producing results that can be applied to existing seamless cloud services linking essential mission data from multiple sources to facilitate automated weapon planning processes. GBL will leverage its prior SBIR and commercial technologies to reduce development risk and increase the feasibility ofsuccessful extensibility and transition of WAAP software products.