Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Jul 4, 2019
In this video, we show the flexibility of our modular framework for motion planning to utilize various high-level planning algorithms. The high-level plan is realized at the continuous time level by feedback-based motion primitives, with no further refinements. As we show, different algorithms are more suitable to different scenarios that contrast in the number of vehicles, whether they move simultaneously, the nature of the task, and the overall computation time. We compare a non-deterministic Dijkstra algorithm (for moving vehicles simultaneously and optimally), deterministic A* (single vehicle moves at a time and optimally with faster computation), and deterministic greedy search (single vehicle moves at a time, not necessarily optimally but with very fast computation).