Obstacle detection (OD) for autonomous ground vehicles on highways and in dynamic environments is exceptionally difficult due to the short response time requirements. Many OD systems handle this problem by using sensors with less data or by using powerful processors. However, this problem can be solved using biology-inspired processing without replacing hardware. LiDAR is one of the most common sensor options for OD in autonomous ground vehicles. The importance map, a recently developed biology-inspired processing technique using events, has been able to efficiently highlight LiDAR returns in a scene that correspond to obstacles. Even though the importance map has shown potential, there are three significant flaws: no object movement distinction, high levels of noise, and poor static object tracking. Solutions for these flaws are presented in this research. The constant-angle principle identifies motion toward the ego vehicle, temporal filtering removes noise, and an updated static object-tracking algorithm performs well at various speeds. OD capabilities of the importance map and the previous importance map are compared using LiDAR data from the KITTI data set. Using comparisons between true-positive and false-positive rates, the importance map performs much better than the previous importance map. Furthermore, events are shown to be a highly discriminative feature for finding obstacles in a scene.