On-board Autonomous Surveillance and Navigation System

The Army seeks to develop a system to allow an unmanned ground vehicle (UGV) to extract useful information from surveillance video, to reposition itself to optimize data collection and communications, and to conceal itself. To meet the Army’s requirement, BCI proposes an Onboard Autonomous Surveillance and Navigation System (OASN) with Enhanced Accuracy and Low Computation Cost for UGVs, based on our extensive experience in target detection and recognition, intelligent video analysis algorithms, IR/visual sensor developments, and autonomous navigation systems for ground vehicles. BCI’s proposed OASN efficiently integrates our two innovations—ultra-fast human and vehicle activity auto-detection and high-accuracy object-oriented video navigation—into a compact, onboard video surveillance system, which not only achieves ultra-fast and highly accurate human and vehicle detection over in-the-field UGV platforms, but also supports autonomous UGV and sensor position adjustments and concealment improvement. The proposed algorithms are of very low computation cost and support ultra-fast initial system setup and configuration. The OASN algorithms are also energy efficient for onboard calculation. BCI’s proposed novel OASN system can be easily implemented with the commercial-off-the-shelf (COTS) products and open source products by means of flexible hardware and software design, and is capable of working with any UGV platform.