We can easily obtain cheap off-the-shelf microprocessors, wireless radios, and sensors. A variety of actuators from motors and servos all the way to fully built toy vehicles are also readily available. We can build wireless sensor networks from the former and attach them to mobile mechanical bodies from the latter. So where are our autonomous networked swarm robots then?
It turns out there are still a number of system-level problems to solve before we can have personal robots to do our bidding. In this talk, we look at a few of these issues -- how they impede the development of swarm robots, and what can be done to solve or work around them. These range from networking concerns such as power consumption and integration to robotic concerns including localization and control. At the interface are further concerns regarding communicating with mobile nodes. This presentation will cover protocol design and enhancements for networking along with hardware and software development for robot control, with a few detours and asides on the way. We finish by presenting plans for how to move forward so that we can eventually have our own versatile army of robot servants.