Integrated Perception and Planning for Autonomous Navigation through Vegetated Terrains

Project Team

Principal Investigator

Christopher Goodin, Center for Advanced Vehicular Systems, Mississippi State University Tulga Ersal, University of Michigan Bogdan Epureanu, University of Michigan

Government

Paramsothy Jayakumar, U.S. Army GVSC

Faculty

Ethan Salmon, Mississippi State University

Industry

Brittney English, Dynetics

Chenyu Yi, Mercedes-Benz

Andrew Kwas, Timothy Morris, Northrop Grumman

Student

Marc Moore, Riku Kikuta, Mississippi State University

James Baxter, Rishitha Paga, Junsik Eom, University of Michigan

Project Summary

Project IE.03 began in 2024 and was completed June 2025.

This integration effort produced the case study “Forging Further with Foliar Foresight: Perception and Planning for Autonomous Navigation through Vegetated Terrains” presented at the 2025 ARC Annual Program Review.

Case Study Abstract

Autonomous off-road vehicles face the critical challenge of navigating complex terrains characterized by varying topologies and vegetation. This integration effort enhances autonomous offroad vehicle mobility by developing an advanced trajectory planner that incorporates both terrain topology and vegetation resistance into its navigation strategies.

This work builds on two ongoing ARC projects. Project 1.40 has introduced a touch-based sensor on an MRZR platform to characterize vegetation override forces. This integration effort extends this work by training a machine learning model to map these forces using aerial imagery. Meanwhile, Project 1.41 has developed a trajectory planner for highly-mobile navigation on difficult terrains. This integration effort further extends this planner to include considerations for vegetation resistance.

The newly integrated framework is implemented and tested on Mississippi State University’s MRZR and proving grounds. Simulation studies provide additional validation. Results demonstrate a significant improvement in autonomous off-road mobility compared to planners that ignore vegetation resistance and rugged terrain topology. Thus, this integration effort demonstrates unprecedented capabilities for autonomous off-road mobility.

Off-road navigation involves balancing a variety of tradeoffs. One tradeoff is demonstrated in the work of Project #1.41, where the improved grade-climbing gained by approaching slopes at higher speeds must be balanced against the inherent risk of cresting a hill and traveling into terrain beyond the current line-of-sight at that high speed. The situation is similar for traversing through vegetated terrain. Natural terrains typically progress from small, dense grasses to larger, sparser trees. In the first year of Project #1.40, measurements revealed that even the thickest grasses offer far less override resistance than sparse small trees and shrubs. However, these thick grasses also have much lower visibility range for sensors like Lidars and cameras. Therefore, navigation through vegetation also represents a tradeoff between areas with lower resistance but higher uncertainty about the terrain or areas with higher mobility resistance but less uncertainty about the terrain. In this integration project, we will develop an integrated perception and planning system that can balance these competing objectives.

The goal of this integration project is to develop an integrated perception and planning framework for off-road navigation that balances the tradeoff between maximizing mobility through slopes and vegetation and minimizing risk for the vehicle. This framework comprises an extension of the local trajectory planner developed in Project #1.41 with the real-time vegetation classifier developed in Project #1.40 to estimate the motion resistance induced by vegetation. The framework also includes a novel map-level vegetation override model.

IE.03