Dynamic Target Tracking with Re-ID for Autonomous Terminal Guidance

As of late, massive investments and open-source development of AI and computer-vision technologies has served as catapult for major advancements in detection, recognition, classification, and tracking algorithms.Ā Still, many statistically based detection and tracking algorithms are computationally expensive and require large compute resources to operate real-time, leading to compromises in execution methodologies and performance. In addition, required processing resources limit the ability to deploy such methods on platforms with stringent cost, size, weight and power (C-SWaP) constraints, such as Group 2+ and below UAS. Further computational heavy pipelines may significantly reduce platform power endurance and associated mission effectiveness. As such, alternative algorithms must be developed with similar, if not improved, performance but requiring significantly less computational resources. Lastly, any new or deployed model must not be reliant on massive libraries of training data that may not be relevant or can easily be tricked by simple camouflage or other shape changes to target platforms. OKSI will implement an autonomous target tracking and autonomous terminal engagement capability for sUAS platforms that have limited compute capabilities that can readily be integrated onto a range of sUAS systems that are presently employed throughout the battlefield as a software upgrade.