March 2006, Issue 188

Robot Localization and Control

Imagine you’re in a grocery store comparing the ingredients in different brands of applesauce. As you’re reading one of the labels, the jar slips from your hand and shatters on the floor in a miniature explosion of glass shards and apple puree. Looking over the damage, you begin to contemplate whom you should inform about the mess, when suddenly, a small wheeled robot rounds the corner and makes a beeline toward the blast site

Intrigued, you watch as the robot seems to pause and assess the situation. Two more robots then arrive on the scene. One efficiently scoops up the mess, and the other wipes the floor clean. The first robot monitors their progress. After a few minutes, the mess is gone, along with the robots, and the store’s management has been informed of its applesauce casualty.

Such a team of robots is the result of an area of research that deals with autonomous robot teams and swarm robotics. This interesting field covers problems that involve a diverse network of specialized robots sent out to accomplish a variety of specific tasks. In the grocery store example, one robot is a first-response unit that determines which kind of job is to be done. After surveying the situation, it calls on two other robots to remove the mess and wipe the floor clean. Additional robots in the grocery store network might specialize in different cleaning tasks, restocking shelves, and even leading customers around the store.

One of the hurdles to implementing this kind of robotic team is the problem of localization, which is a key part of any mobile robotic network. Localization is the process of determining the location of nodes within the network as accurately as possible. In this article, we’ll present a ZigBee-based localization solution for estimating a robot’s position.