SBIR-STTR Award

Integrated Adaptive Applications (IAA), Inc., in collaboration with Delphi Automotive Systems, LLC, proposes a networked automotive radar processing system. The system utilizes multiple Delphis mid-range radar (MRR) units to enhance the capabilities of each individual MRR unit, while still maintaining the low-cost advantage. Specifically, by deploying multiple MRR units in a networked fashion, a forward-looking system, for example, can have a virtual array with an aperture length that is up to twice the front bumper length, making it possible to achieve an extremely fine azimuth angular resolution. The significantly improved azimuth discrimination capability allows for much improved object sizing and classification abilities, as well as for enhanced detection range. We further propose to devise a robust high-resolution direction-of-arrival (DOA) estimation method and a self-calibration method. The proposed DOA estimation method can achieve further improved angular resolution while significantly suppressing the sidelobe levels. It works with few or even a single snapshot, and is robust against the challenging source-coherence problem, which causes many array processing algorithms in the literature to fail completely. The self-calibration method can refine and update the array calibration weights automatically without the requirement of any additional hardware or the need of interrupting the radar operation.

Integrated Adaptive Applications (IAA), Inc., in collaboration with Delphi Automotive Systems, LLC, proposes to develop a networked automotive radar processing system to enhance the capabilities of each individual MRR unit, at the same time, maintain the low-cost advantage. In the Phase I effort, a network has been proposed using two Delphis mid-range radar (MRR) units. The network can quadruple the array aperture size and spatial degrees of freedom, compared to 1 MRR unit only. It has been shown by numerical examples that the azimuth angle resolution can be improved from 16 degrees to 1.2 degrees by using the proposed network, together with an iterative adaptive angle estimation approach. The objective of the Phase II program is to mature the 2-unit MRR network via hardware development, and to demonstrate their technical feasibility using both simulated and field collected data. First, a hardware/software mixed prototype system will be developed for fielded data collection and demonstration. Second, the 2-unit MRR network and the associated signal processing algorithms will be validated with both simulated and field collected data, and their benefits will be clearly demonstrated. Third, a clear path to real-time implementation of the proposed algorithms will be demonstrated.