Search Results

Autonomous driving has witnessed many developments in the past decade. Most automotive manufacturers and technology companies have been investing in the software and platform development of these vehicles, and long strides have been made in this domain. Nonetheless, academia has been partially isolated from the practical efforts of these developments, due to the cost involved in converting a standard vehicle into a drive-by-wire platform suitable for self-driving vehicular research. In this paper, a low-cost autonomous vehicle platform is developed and the process for converting a Polaris Gem e2 FMVSS-500 vehicle to brake-by-wire, throttle-by-wire, and steer-by-wire is elaborated. The tools and the procedures followed here were all conducted by students, in house, and can be easily duplicated for other FMVSS-500 compatible vehicles.

A novel multi-modal scene segmentation algorithm for obstacle identification and masking is presented in this work. A co-registered data set is generated from monocular camera and light detection and ranging (LIDAR) sensors. This calibrated data enables 3D scene information to be mapped to time-synchronized 2D camera images, where discontinuities in the ranging data indicate the increased likelihood of obstacle edges. Applications include Advanced Driver Assistance Systems (ADAS) which address lane-departure, pedestrian protection and collision avoidance and require both high-quality image segmentation and computational efficiency. Simulated and experimental results that demonstrate system performance are presented.