SOFT-BOUNDARY BASED PATH OPTIMIZATION FOR COMPLEX SCENES FOR AUTONOMOUS DRIVING VEHICLES

摘要

According to an embodiment, an ADV includes a method to plan a path. The method selects a plurality of discrete points of a path to determine the obstacle cost for the path. For the discrete points that do not overlap with an obstacle, the method accumulates an obstacle cost of the path using a first cost function starting from the initial discrete point through the discrete points that are successively further away from the initial point until a first discrete point is determined to overlap with an obstacle. The method calculates an overlapping obstacle cost for the first discrete point using a second cost function. The second cost function is based on a distance of the first discrete point from the initial point of the path. The method accumulates the overlapping obstacle cost to the obstacle cost to determine the obstacle cost for the path that overlaps with an obstacle.