This paper focuses on the development of automotive sensors that can measure the relative position and velocity of another vehicle in close proximity, so as to enable prediction of an imminent collision just before the collision occurs. Anisotropic magnetoresistive (AMR) and sonar sensors are adopted for development of the proposed sensor system. The challenges in the use of the AMR sensors include their nonlinear behavior, limited range and magnetic signature levels that vary with each type of car. An adaptive filter based on the extended Kalman filter (EKF) is developed to automatically tune filter parameters for each encountered car and reliably estimate car position. The utilization of an additional sonar sensor during the initial detection of the encountered vehicle is shown to highly speed up the parameter convergence of the filter. Experimental results are presented from a large number of tests with various vehicles to show that the proposed sensor system is viable. The developed sensors represent perhaps the first ever system that can measure relative vehicle position at close proximity right up to the point where a crash occurs.