CHAPTER 4
 COLLECTIVE TARGET TRACKING

 The self-organization algorithm for spreading the robots uniformly over any region

has been discussed and analyzed in Chapter 3. The next goal of moving the robots

collectively towards a target is taken for discussion in this chapter.

 4.1 Requirements and Suitable Modifications

 Application of target tracking algorithm demands the understanding of the proper

environment where it is being applied. According to the requirements of the intended

usage, the mode of application of the algorithm needs to be modified. Below are some of

the different situations encountered in the target tracking scenario.

 The first and the simplest situation is the one where the robots know their own

positions as well as the position of the target. Then, the solution becomes trivial.

 Secondly, a centralized approach may be designed in which one leader robot knows

the position of the target and it guides the group towards it. As explained in Chapter 1,

centralized strategy can have many disadvantages and hence, not a preferred one. One of

the ways this technique can be improved is by having more than one leader in the group

and periodically rotating the leader in a move to make the group robust to failures.

 In some applications, the target itself issues a beacon signal or it reflects signals

from the robots, intimating them on the direction in which it is located. Eisler et. al.

[Eis02] have analyzed the issues related to the tracking of unknown acoustic wave

reflecting objects using a swarm of pressure sensors. In their technique, an emitter sends

an acoustic pulse and the object reflects this pulse. The swarm of mobile sensors receives