Artificial immune systems: application to autonomous agents

The function of the immune system is to protect the living body against invaders through the use of defensive mechanisms. Some previous researchers have used artificial immune systems (AIS) to solve diverse engineering problems. The purpose of the paper is to apply the AIS technique to a distributed autonomous robotics system (DARS) problem. One of the classic problems in DARS is the dog and sheep problem. The authors try to benefit from the features of the natural immune system in the development of the dog and sheep problem. On the other hand, we find that natural immune systems are sophisticated information processors. They learn to recognize relevant patterns; they remember patterns that have been seen previously; and, they use diversity to promote robustness. Furthermore, the individual cells and molecules that comprise the immune system are distributed throughout the body, encoding and controlling the system in parallel, with no central control mechanism. The immune system uses several weapons to attack the foreign antigen. Abstractly, these weapons are the helper T-cells, B-cells, and antibodies. We simulated the dog as a B cell, the sheep as an antigen, the antibody as the dog behavior, the antigen response as the sheep behavior, and the sheep-to-pen distance as a helper T cell. The system interacts in an equivalent manner to the immune response, trying to restore the environment to its original state, which is the sheep inside the pen.