Robust On-line Obstacle Detection using Range Data for Reactive Navigation

https://doi.org/10.3182/20120905-3-HR-2030.00130Get rights and content

Abstract

This paper proposes a robust on-line and adaptive elliptic trajectory for reactive obstacle avoidance. These trajectories permit a safe and smooth mobile robot navigation in cluttered environment. Indeed, they use limit-cycle principle already applied in the literature Adouane et al. (2011). The main contribution proposed here is to perform this navigation in a completely reactive way while using only uncertain range data. Each obstacle, in this obstacle avoidance strategy, is surrounded by an ellipse and its parameters are obtained online while using the sequence of uncertain range data. This method uses the fusion between heuristic approach and Extended Kalman Filter (EKF) techniques to improve the computed ellipse parameters. A large number of simulations and experiments show the efficiency of the proposed on-line navigation in cluttered environment.

Keywords

Mobile robots navigation
Obstacle detection and avoidance
Telemetry
Parameter identification
Extend Kalman Filter

Cited by (0)

Supported by the French National Research Agency (ANR) through the Safeplatoon and R-Discover projects.

View Abstract