Collision Avoidance Radar: Automotive

In this study the application of millimeter-wave radar systems as a remote sensing tool in identifying and warning of the presence of hazardous conditions in a highway environment is investigated. Millimeter-wave radars, whose signal is capable of penetrating through fog, snow, and rain, are most suitable for autonomous vehicle control operations because of their ability in measuring the target range. Polarimetric backscatter measurements from different distributed targets, such as asphalt, gravel, ice-covered asphalt, etc., and point targets, such as potholes, bricks, tires, road-signs, etc., are be conducted using The University of Michigan's 35 and 94 GHz radar systems. Measurements will be conducted at incidence angles ranging between 70° and 88° from nadir. Such data set which is essential for the design of collision warning and other radar-based sensors for automotive applications does not exist. In many radar applications, detection and discrimination of a target in the presence of clutter can be enhanced significantly by appropriately choosing the optimal set of transmit and receive polarizations. The ongoing investigation will determine the feasibility of low radar cross section target detection on the road surface using the optimal polarization scheme. In characterization of the optimal polarization it is shown that the objective function of is highly nonlinear and discontinuous, hence classical optimization algorithms fail to provide satisfactory results. A genetic algorithm which operates on a discretized form of the parameter space and searches globally for the optimum point will be used. The set of polarimetric backscatter measurements of asphalt surfaces under different physical conditions will be used to come up with the optimal design for polarization states of an affordable millimeter-wave radar sensor that can assess traction of road surfaces.