Dynamic User Task Scheduling for Mobile Robots

We present our efforts to deploy mobile robots in office environments, focusing in particular on the challenge of planning a schedule for a robot to accomplish user-requested actions. We concretely aim to make our CoBot mobile robots available to execute navigationalbased tasks requested by users, such as telepresence, and picking up and delivering messages or objects at different locations. We contribute an efficient webbased approach in which users can request and schedule the execution of specific tasks. The scheduling problem is converted to a mixed integer programming problem. The robot executes the scheduled tasks using a synthetic speech and touch-screen interface to interact with users, while allowing users to follow the task execution online. Our robot uses a robust Kinect-based safe navigation algorithm, moves fully autonomously without the need to be chaperoned by anyone, and is robust to the presence of moving humans, as well as non-trivial obstacles, such as legged chairs and tables. Our robots have already performed 15km of autonomous service tasks. Introduction and Related Work We envision a system in which autonomous mobile robots robustly perform service tasks in indoor environments. The robots perform tasks which are requested by building residents over the web, such as delivering mail, fetching coffee, or guiding visitors. To fulfill the users’ requests, we must plan a schedule of when the robot will execute each task in accordance with the constraints specified by the users. Many efforts have used the web to access robots, including the early examples of the teleoperation of a robotic arm (Goldberg et al. 1995; Taylor and Trevelyan 1995) and interfacing with a mobile robot (e.g, (Simmons et al. 1997; Siegwart and Saucy 1999; Saucy and Mondada 2000; Schulz et al. 2000)), among others. The robot Xavier (Simmons et al. 1997; 2000) allowed users to make requests over the web for the robot to go to specific places, and other mobile robots soon followed (Siegwart and Saucy 1999; Grange, Fong, and Baur 2000; Saucy and Mondada 2000; Schulz et al. 2000). The RoboCup@Home initiative (Visser and Burkhard 2007) provides competition setups for indoor Copyright © 2011, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Figure 1: CoBot-2, an omnidirectional mobile robot for indoor users. service autonomous robots, with an increasingly wide scope of challenges focusing on robot autonomy and verbal interaction with users. In this work, we present our architecture to effectively make a fully autonomous indoor service robot available to general users. We focus on the problem of planning a schedule for the robot, and present a mixed integer linear programming solution for planning a schedule. We ground our work on the CoBot-2 platform 1, shown in Figure 1. CoBot-2 autonomously localizes and navigates in a multi-floor office environment while effectively avoiding obstacles (Biswas and Veloso 2010). The robot carries a variety of sensing and computing devices, including a camera, a Kinect depthcamera, a Hokuyo LIDAR, a touch-screen tablet, microphones, speakers, and wireless communication. CoBot-2 executes tasks sent by users over the web, and we have devised a user-friendly web interface that allows users to specify tasks. Currently, the robot executes three types of tasks: a GoToRoom task where the robot visits a location, a Telepresence task where the robot goes to a location CoBot-2 was designed and built by Michael Licitra, [email protected], as a scaled-up version of the CMDragons small-size soccer robots, also designed and built by him. Server side Robot side interface agent executing manager agent