Potential field-based mobile node movement

A sensor network contains a set of mobile nodes. The idea is that the nodes are moving towards the areas where there is interesting phenomena (but, also, they are trying to load-balance themselves, that is, not gather all of them to the hottest part and ignore the rest). Design a mobility pattern based on an artificial potential fields. The assumption is to assume that there are some artificial forces acting in the field. Areas of interest are attracting the nodes, while the nodes are attracting each other from longer distance, but are having a repulsion on small distance to prevent them clustering together. Discussion Potential field-based approaches are quite popular in mobile robotics. One paper to read is JH Reif, H Wang "Social potential fields: A distributed behavioral control for autonomous robots" . But many other systems use similar approaches, although they might not call them social potential fields (for instance, many of the Robocup competitors). Here are some things you might want to try out:

• The SPF paper uses inverse polynomial forces. These are like gravitation: given the right initial speed, the node will not converge to the zone of the attraction, but take up an orbit around it! You might want to try out forces with different expressions. For instance you can introduce friction (a constant force acting against the movement), viscosity (a force acting against the movement, proportional with the speed). Or you can make the forces elastic forces - proportional with the distance - that is, the farther you are, the stronger the attraction.
• The other idea is to actually exploit the properties of the inverse polynomial forces. Maybe you do want your sensor to take up orbit around the area of interest - especially if your sensors are mounted on unmanned aerial vehicles (UAVs).

MORE INFO TO COME