# 8672 / Trajectory optimization for vegetation override in off-road driving Title: Trajectory optimization for vegetation override in off-road driving Authors: Charles Noren, Bhaskar Vundurthy, Sebastian Scherer, and Matthew Travers Abstract: Off-road robotic vehicles often encounter environmental objects while traversing complex terrains. These objects may obstruct the vehicle’s motion and must be perceived and reasoned about to ensure safe operation. Historically, limited sensing fidelity for autonomous vehicles led to either overly cautious or unnecessarily risky behaviors when interacting with environmental objects. Some objects must be avoided to ensure operational integrity, but avoiding all environmental objects will likely cause needlessly slow traversals. With contemporary sensing, the capability to tell the difference between these environmental objects is now possible. Given this improved sensing capability, developing controls algorithms which not only reason about avoiding certain classes of objects but can also reason about motions through other classes of objects is now the next step towards robust off-road autonomous driving. In this work, we propose a trajectory optimization technique for overriding vegetation that explicitly encodes contact constraints for a subset of vegetative objects found in the operating environment. The subset of objects that the vehicle can safely collide with is derived from existing parameterized collision models, the parameters of which are estimated by the on-board sensing system, and then characterized by a necessary velocity at the point of collision. These constraints are then enforced during vehicle run-time by the controller. The controller's capabilities are shown in both simulation and used to control a full-sized autonomous utility task vehicle. The vehicle’s determination to avoid or override objects is robust to uncertainty in the sensing system and is dependent on the specifics of the environment and object geometry. Order the full paper: ISTVS members: receive three papers per year as part of your membership via the ISTVS Member Portal: --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://2023.istvs.org/submissions/papers/8672.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.