Automated Target Tracking -- Closing the gap between the Predictable and the Unpredictable

In the world of autonomous control of a UAV, there are many things that can be predicted and accounted for. The path that a vehicle will travel can be mapped from known navigable routes. Static objects that box that path can be identified and avoided using long-range imaging, plus non-real-time processing and adaptive path planning, which can be a combination of onboard and remote monitoring assets. Other moving vehicles can be made cooperative in their operation via rules of operation and telemetry transmitting devices. All of these things create an environment whereby location technology and rules execution can enable successful autonomous vehicle operation.

This paper explores the ramifications of real-time autonomous vehicle performance in cluttered environments. It analyzes the gap between the predictable and the unpredictable, introduces real object tracing, explores the benefits and pitfalls of real-time moving object tracking and introduces possible solutions for uninterrupted operation of autonomous vehicles.