A New Belly-Dancing Robot

By Roland Piquepaille

She can shimmy, she can roll, she can backbend. She even sports a teasing, low-slung skirt around her waist. But the performer of these undulations is no fleshy temptress. Instead, she is a belly-dancing robot whose moves are driven by the wriggles of a fish.

Here is a picture of Jimmy Or's belly-dancing robot (Copyright Jimmy Or).

Now, let's look at some technical details.

To generate the robot's undulations, Or borrowed a computer program built by Swedish researchers that simulates a network of nerves in the lamprey called a central pattern generator (CPG). The CPG directs the lamprey's movements without the help of the brain or sensory feedback.

Similar nerve networks are thought to exist in most vertebrates. In chickens, it is a CPG that allows a headless bird to briefly sprint around the yard before keeling over, for example. In humans, a CPG is thought to produce an automatic walking motion in toddlers or people who have had spinal-cord injuries when they are placed on a treadmill.

Here is Or's conclusion.

Or admits that he made Waseda Belly Dancer No. 1 partly as an entertainment; he is currently refining its workings and choosing some fetching jewellery for it. But he maintains that robots with a flexible spine have a future. "A robot that can bow is very important in Japanese society," Or says.

Back when he was at the University of Edinburgh, Or published with some of his colleagues a technical paper about the lamprey, "Evolution of efficient swimming controllers for a simulated lamprey." Here is the abstract.

This paper investigates the evolutionary design of efficient connectionist swimming controllers for a simulated lamprey. Efficiency is defined as the ratio of forward swimming speed to backward mechanical wave speed.Using the lamprey model proposed by Ekeberg (1993) and extending the work of Ijspeert et al. (1999) on evolving lamprey swimming central pattern generators (CPGs) through genetic algorithms (GAs), we investigate the space of possible neural configurations which satisfies the property of high swimming efficiency. Techniques are devised to measure efficiency at various swimming speeds. The measurements are incorporated into the fitness function of Ijspeert's original GA and efficient controllers are evolved. Interestingly, the best evolved controller not only is capable of swimming in a similar manner to the real lamprey, but also with the same efficiency (about 0.8). Moreover, it can exhibit a wide range of controllable speeds and efficiencies.

Source: Helen Pearson, Nature, January 20, 2004