A robust set approach for mobile robot localization in ambient environment

Abstract

Mobile robot localization consists in estimating the robot coordinates using real-time measurements. In ambient environment context, data can come both from the robot on-board sensors and from environment objects, mobile or not, able to sense the robot. The paper considers localization problem as a nonlinear bounded-error estimation of the state vector. The components of the state vector are the robot coordinates as well as the 2D position and orientation. The approach based on interval analysis can satisfy the needs of ambient environment by easily taking account a heterogeneous set and a variable number of measurements. Bounded-error state estimation can be an alternative to particle filtering which is sensitive to non-consistent measures, large measure errors, and drift of evolution model. The paper addresses the theoretical formulation of the set-membership approach and the application to the estimation of the robot localization. Additional treatments are added to the estimator in order to meet more realistic conditions. Treatments aim at reducing the effects of disruptive events: outliers, model inaccuracies or model drift and robot kidnapping. Simulation results show the contribution of each step of the estimator.

Probability Dueling DQN active visual SLAM for autonomous navigation in indoor environment

“…In indoor environments, mobile agents need to be capable of autonomous navigation, obstacle avoidance and mapping. Simultaneous localization and mapping (SLAM) Colle and Galerne (2018) is an algorithm for real-time localization and mapping using information obtained by agents through sensors. There have been many related research results in past decades.…”

“…Autonomy in the complex environment is very important to mobile robot, which requires the robot can navigate autonomously, avoid obstacles and mapping. Localization and mapping have been well-studied in the past decades based on simultaneous localization and mapping (SLAM) [1]. However, a path planning algorithm integrated with SLAM to form an all-encompassing framework is still in its early years.…”