Evidence-Based Landing Site Selection and Vison-Based Landing for UAVs in Unstructured Environments

2026-05-02

AI Summary

This work addresses the challenge of safe autonomous landing for unmanned aerial vehicles in unstructured environments, where perception noise and platform disturbances compromise reliability. The authors propose an evidential probabilistic framework that decouples decision-making under uncertainty from visual servoing. By recursively fusing frame-level visual cues, the method constructs a spatiotemporally consistent safety belief map, augmented with hard geometric constraints—such as size, slope, and flatness—to eliminate infeasible landing regions. The final landing site is selected via a constrained maximum a posteriori estimation, and precise touchdown is achieved through ORB feature tracking coupled with image-based visual servoing (IBVS). A key innovation lies in modeling landing safety as a latent variable and employing evidence accumulation for robust belief updating. Experiments in real-world settings and NVIDIA Isaac Sim simulations demonstrate the approach’s stability, consistency, and strong robustness to perceptual errors.

Abstract

Autonomous landing in cluttered or unstructured environments remains a safety-critical challenge for unmanned aerial vehicles (UAVs), particularly under noisy perception caused by sensor uncertainty and platform-induced disturbances such as vibration. This paper presents an evidence-based probabilistic framework for autonomous UAV landing that explicitly separates decision-making under uncertainty from execution via visual servoing. Landing safety is modeled as a latent variable and inferred through recursive accumulation of frame-wise visual likelihoods derived from flatness, slope, and obstacle cues, yielding a temporally consistent belief map that is robust to transient perception errors. Physical feasibility is enforced through a hard geometric constraint based on the minimum required landing radius of the UAV, ensuring that undersized but visually appealing regions are rejected. The final landing site is selected using constrained maximum a posteriori estimation. Once selected, the UAV locks onto the target region using ORB feature tracking and performs precise alignment and descent via image-based visual servoing (IBVS). The proposed approach is validated through both real-world laboratory experiments and high-fidelity simulations in Nvidia Isaac Sim, demonstrating consistent, cautious, and stable landing behavior across domains.

Problem

• autonomous landing
• unstructured environments
• sensor uncertainty
• perception noise
• UAV safety

Innovation

• evidence-based landing
• visual servoing
• uncertainty-aware decision-making
• belief map
• geometric feasibility constraint