Local optimization of cooperative robot movements for guiding and regrouping people in a guiding mission

Abstract

Abstract-This article presents a novel approach for optimizing locally the work of cooperative robots and obtaining the minimum displacement of humans in a guiding people mission. Unlike other methods, we consider situations where individuals can move freely and can escape from the formation, moreover they must be regrouped by multiple mobile robots working cooperatively. The problem is addressed by introducing a "Discrete Time Motion" model (DTM) and a new cost function that minimizes the work required by robots for leading and regrouping people. The guiding mission is carried out in urban areas containing multiple obstacles and building constraints. Furthermore, an analysis of forces actuating among robots and humans is presented throughout simulations of different situations of robot and human configurations and behaviors.

Continuous real time POMCP to find-and-follow people by a humanoid service robot

“…Next we would like to use a smarter prediction of the person's path, instead of a random movement [5]. Furthermore the adaptive method could be improved by giving preference to close search locations.…”

“…In this way we can do real-time experiments with human volunteers. Moreover, we compare our methods with a previous and well-known system: heuristic person follower [5] Through a large number of synthetic and real experiments, we observed that the Adaptive Highest Belief Continuous Real-time POMCP Follower was able to find and follow different volunteers, while the other two methods performed the task slowly or lost the person, and therefore could not continue the mission.…”

Interactive multiple object learning with scanty human supervision

“…Computer vision techniques for human-robot interaction have been mainly focused on recognizing people in urban scenarios [10,34,39] as well as identifying human gestures and activities [13,36] to establish contact with people and perform particular robotics tasks such as guiding people in museums and urban areas [18,42,49], providing information in shopping malls [23], or recognizing human emotions through classifying facial gestures [9]. Although these techniques have endowed the robot with remarkable interaction skills, they are commonly computed offline and using a potentially large training time.…”

Cooperative social robots to accompany groups of people

“…In a previous work (Garrell and Sanfeliu, 2010a), we studied how robots should behave when they accompany groups of people, or which strategy the group of robots should follow when a person moves away. In this work, we go one step further, presenting a model designed to prevent people from moving away from the formation.…”

“…The results, a discussion of the methods used and conclusions are presented in Sections 8, 9 and 10 respectively. This paper is an expanded version of the work presented in (Garrell and Sanfeliu, 2010a). In this current version we have included a new model called Prediction and Anticipation Model which has been developed to predict human behavior and prevent people from straying from the formation.…”