SBIR-STTR Award

SLITHER: Snake-Like Inspection Technology for Hazardous and Enclosed Regions
Award last edited on: 4/22/2026

Sponsored Program
SBIR
Awarding Agency
DOD : AF
Total Award Amount
$1,249,999
Award Phase
2
Solicitation Topic Code
AF254-D0824
Principal Investigator
Matt Brown

Company Information

Thoughtforge Inc

1329 66th Street Apt C
Berkeley, CA 94702
   (415) 975-1369
   N/A
   www.thoughtforge.ai
Location: Single
Congr. District: 12
County: Alameda

Phase I

Contract Number: N/A
Start Date: 11/4/2025    Completed: 11/4/2027
Phase I year
2026
Phase I Amount
$1
Direct to Phase II

Phase II

Contract Number: FA8571-26-C-0008
Start Date: 11/4/2025    Completed: 11/4/2027
Phase II year
2026
Phase II Amount
$1,249,998
ThoughtForge, Inc. proposes to develop and demonstrate SLITHER (Snake-Like Inspection Technology for Hazardous and Enclosed Regions), a modular, autonomous robotic system designed to address the persistent challenges of inspecting confined, hazardous spaces within aircraft wing fuel tanks on U.S. Air Force platforms such as the C-130, C-5, and C-17. Current manual inspection methods in these environments are labor-intensive, hazardous, inconsistent, and require significant aircraft downtime. SLITHER overcomes these limitations by integrating the FLX BOT—a proven, snake-like robotic platform—with ThoughtForge’s proprietary NetAIF active inference-based control architecture, which enables real-time adaptive behavior based on environmental feedback and sensor input.This R&D effort will focus on developing autonomous navigation, macro- and micro-inspection capabilities, and 3D model generation within complex internal geometries. The system will utilize multi-modal sensing—including visual and depth sensors—to capture high-resolution data, build accurate three-dimensional reconstructions, and ensure inspection completeness. Importantly, SLITHER’s control system is designed to gracefully handle uncertainties such as occlusions, low-light environments, or feature-degraded areas, using active re-localization strategies and fallback navigation modes. Areas that cannot be accessed autonomously will be automatically flagged for human follow-up, ensuring no critical regions are left uninspected. The project also includes the development of an intuitive, interactive Human-Machine Interface (HMI), enabling inspectors to view inspection data, annotate findings such as corrosion, cracks, or FOD directly onto the generated 3D models, and generate permanent, traceable digital inspection records. This user-centered design will be validated through iterative usability testing, preferably in collaboration with Air Force maintenance personnel, to ensure operational relevance and effectiveness. Comprehensive system validation—including autonomous operation, coverage mapping, micro-inspection targeting, and operator interaction—will be conducted using physical mock-ups and simulated environments replicating aircraft wing fuel tanks. The result will be a Technology Readiness Level (TRL) 7 prototype capable of deployment for operational evaluation at Air Force maintenance depots.