Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Occlusion aware obstacle prediction using people as sensors

Sithija Ranaraja

TL;DR

This work tackles the difficulty of perceiving obstacles hidden behind occlusions by exploiting humans as dynamic sensors. It introduces a three-tier approach—Noisy Prediction, Sensor-Based Clearing, and Kalman Filter-Based Fusion—to predict para-occluded obstacles and represent them as temporally annotated 2D Gaussians. Real-time LiDAR clearing and multi-agent fusion reduce false positives and uncertainty, while timestamps guide how predictions should decay or be acted upon, mitigating robot freezing in tight spaces. Across reaction-based and crowd-simulation setups, the framework demonstrates improved occlusion prediction accuracy, lower collision risk, and enhanced navigation efficiency, underscoring its potential for safe operation in crowded, unstructured environments.

Abstract

Navigating dynamic and unstructured environments poses significant challenges for autonomous robots, particularly due to the uncertainty introduced by occluded areas. Conventional sensing methods often fail to detect obstacles hidden behind occlusions until they are dangerously close, especially in crowded spaces where human movement and physical barriers frequently obstruct the robot's view. To address this limitation, we propose a novel framework for occlusion-aware obstacle prediction using people as sensors, that infers the presence of para-occluded obstacles by analyzing human behavioral patterns. Our approach integrates sensor fusion, historical trajectory data, and predictive modeling to estimate the likelihood of obstacle presence and occupancy in occluded regions. By leveraging the natural tendency of humans to avoid certain areas, the system enables robots to proactively adapt their navigation strategies in real time. Extensive simulations and real-world experiments demonstrate that the proposed framework significantly enhances obstacle prediction accuracy, reduces collision risks, and improves navigation efficiency. These findings underscore the potential of occlusion-aware obstacle prediction systems to improve the safety and adaptability of autonomous robots in complex, dynamic environments.

Occlusion aware obstacle prediction using people as sensors

TL;DR

This work tackles the difficulty of perceiving obstacles hidden behind occlusions by exploiting humans as dynamic sensors. It introduces a three-tier approach—Noisy Prediction, Sensor-Based Clearing, and Kalman Filter-Based Fusion—to predict para-occluded obstacles and represent them as temporally annotated 2D Gaussians. Real-time LiDAR clearing and multi-agent fusion reduce false positives and uncertainty, while timestamps guide how predictions should decay or be acted upon, mitigating robot freezing in tight spaces. Across reaction-based and crowd-simulation setups, the framework demonstrates improved occlusion prediction accuracy, lower collision risk, and enhanced navigation efficiency, underscoring its potential for safe operation in crowded, unstructured environments.

Abstract

Navigating dynamic and unstructured environments poses significant challenges for autonomous robots, particularly due to the uncertainty introduced by occluded areas. Conventional sensing methods often fail to detect obstacles hidden behind occlusions until they are dangerously close, especially in crowded spaces where human movement and physical barriers frequently obstruct the robot's view. To address this limitation, we propose a novel framework for occlusion-aware obstacle prediction using people as sensors, that infers the presence of para-occluded obstacles by analyzing human behavioral patterns. Our approach integrates sensor fusion, historical trajectory data, and predictive modeling to estimate the likelihood of obstacle presence and occupancy in occluded regions. By leveraging the natural tendency of humans to avoid certain areas, the system enables robots to proactively adapt their navigation strategies in real time. Extensive simulations and real-world experiments demonstrate that the proposed framework significantly enhances obstacle prediction accuracy, reduces collision risks, and improves navigation efficiency. These findings underscore the potential of occlusion-aware obstacle prediction systems to improve the safety and adaptability of autonomous robots in complex, dynamic environments.
Paper Structure (9 sections, 19 equations, 14 figures, 1 table)

This paper contains 9 sections, 19 equations, 14 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Problem Statement
  4. Methodology
  5. Noisy Prediction Model
  6. Sensor-Based Clearing
  7. Kalman Filter-Based Fusion
  8. Results & Evaluation
  9. Conclusion

Figures (14)

  • Figure 1: Occlusion handling in navigation
  • Figure 3: Gaussian obstacle prediction based on agents' moving direction and clock-wise turning angle
  • Figure 4: Gaussian obstacle prediction based on agents' moving direction and anti-clock-wise turning angle
  • Figure 5: Sensor-based clearing sample configuration
  • Figure 6: Scaler for covariance matrices Based on this scaling factor, any obstacle with higher timestamp will get higher uncertainity in all state variables and gets lower uncertainity when timestep is lower.
  • ...and 9 more figures