Vision-based Monte Carlo localization for RoboCup Humanoid Kid-Size League

Abstract

Localization is the most fundamental ability for winning the RoboCup Humanoid League Competition. In this paper, we present a vision-based localization method called Monte Carlo Localization (MCL) to deal with the limited landmarks left in RoboCup, such as the yellow goal posts and field markers. In the beginning, we give brief explanation of perception system. Next, we give detailed implementation of MCL, an improvement of the resampling step that has been develop before, and the process of estimating the localization result. We perform all experiments on our humanoid robot named Zared_v1.0. Results show that the modified resampling technique in MCL give better result in estimating robot position and orientation on normal and kidnapping condition.

Tiny-YOLO distance measurement and object detection coordination system for the BarelangFC robot

“…Moreover, Monte Carlo localization (MCL) can be considered in order to achieve robot localization. Hartfill [20] developed an MCL scheme based on a 2D RGB image to retrieve the localization information and evaluate it in simulation. Dalmasso et al [21] used a Monte Carlo tree search (MCTS) to decentralize one robot with human to understand the location.…”