traversability

A GPU implementation of Model Predictive Path Integral (MPPI) control that uses a probabilistic traversability model for planning risk-aware trajectories.

Wild Visual Navigation: A system for fast traversability learning via pre-trained models and online self-supervision

[RA-L '24] Learning self-supervised traversability with navigation experiences of mobile robots: A risk-aware self-training approach

WayFASTER: a Self-Supervised Traversability Prediction for Increased Navigation Awareness

off-road navigation simulator for benchmarking planning algorithms

WayFAST: a minimal data waypoints free autonomous navigation algorithm for field robots

This is the official implementation of the ICRA 2025 paper TANGO: Traversability-Aware Navigation with Local Metric Control for Topological Goals

Follow the Footprints: Self-supervised Traversability Estimation for Off-road Vehicle Navigation based on Geometric and Visual Cues

Geometric Traversability using Self-Supervised with Anomaly Classification @ ICRA'26

ASCC2022 - TROT-Q: TRaversability and Obstacle aware Target tracking system for Quadruped robots

A C++ library for 3D traversability estimation from MLS maps

This repository provides a traversability benchmark for experiments in both simulation and real-world environments.

From-scratch C++17 LiDAR + camera perception — sector RANSAC, LiDAR-inertial SLAM, MOT, probabilistic traversability, CBF safety. 13 engineering deep-dives.

[IROS 2026] TravKAN: Fast and Interpretable Nonlinear Traversability Analysis with Kolmogorov-Arnold Networks

Two-stage vision-language traversability analysis for autonomous vehicles using Qwen2-VL sentinel + NVIDIA Cosmos-Reason1 deep reasoning on Jetson AGX Orin.

Multi-task perception network for quadruped robots: traversability segmentation, terrain classification and anchor-free object detection in one PyTorch forward pass, exported to ONNX/TensorRT for Jetson Orin.

Onboard image segmentation for off-road navigation — DeepLabV3+ over 9 traversability classes with a composite CE/Dice/IoU loss and alpha scheduling (Machine Learning course project)

Panoptic segmentation for autonomous forest navigation

Given the geometry of your robot, labels as traversable the points in the point cloud through which its trajectory passes.

Physics-first UGV traversability: Bekker–Wong–Janosi gates, NASA/Wong literature soils, residual adapters, ROS pickup. Research software — not field-certified.