Navigating in populated environments by following a leader

Abstract

Service robots have a great potential of improving human quality of life by aiding in everyday tasks. However, robots that share an environment and interact with humans still face some challenges that limits their acceptance. One of these challenges is how to move and behave among groups of people, which is a task performed seamlessly by humans and some animals. This is an interesting issue for the robotics community, as it is important to predict and adapt to an environment that is constantly changing, while at the same time respect social conventions. Path planning in dynamic environments has been addressed mostly by predicting future position of humans and avoiding them. However, with the increase of the number of persons in such environments, techniques that are based only on the prediction of the movement of humans can fail, as they usually ignore the human's reaction to the presence of the robot. Instead of trying to model the complex human motion behavior, this work proposes to rely on humans to guide the robot through difficult situations, where classical approaches would fail to find a solution. This will be accomplished by a probabilistic approach for electing a human leader, according to the robot's desired destination. In this way, the robot can take advantage of the humans' paths and behavior, effortlessly avoiding dynamic and static features as the human leader does, relieving the robot from the burden of having to generate its own path in difficult situations.

A framework for interactive teaching of virtual borders to mobile robots

“…We chose visual markers because they are widely used for preliminary experiments, e.g. [25], [26], [27], and provide an easy human-robot interface. Since the framework is easily adaptable, it will serve as basis for more intuitive human-robot interfaces in the future.…”

Using contact-based inducement for efficient navigation in a congested environment

“…The tour guide robot Minerva [5] which quickly followed the RHINO experiment, used a similar system while being exhibited at the Smithsonian and during its operation at Few examples of studies that tackle the challenge of crowded navigation include [6,7,8]. In [6], the authors present an algorithm for following a suitable leader in a crowd. [7] describes a situation called 'micro-conflict', and models it as a two player game using supervised learning algorithms.…”

Experiments in Leader Classification and Following with an Autonomous Wheelchair

“…According to these observations, we study the novel hypothesis that a robot can that properly select persons as leaders in a complex environment, will benefit from their advanced navigation and interaction skills, by following them [9]. Figure 1 illustrates the advantages of this approach in comparison with a state of the art motion planning algorithm (RiskRRT), using a pedestrian simulator in a narrow corridor.…”