The biological immune system is a complex, adaptive, pattern-recognition system that defends the body from foreign pathogens. The system uses learning, memory, and associative retrieval to solve recognition issues and classification of tasks. In particular, it learns to recognize relevant problems, remember those encountered in the past, and uses combinations to construct problem detectors efficiently. This paper explores an application of an adaptive learning mechanism for robots based on the natural immune system, using two algorithms, viz., the behavior arbitration mechanism and the clonal selection algorithm to demonstrate the innate and adaptive immune response respectively. The work highlights the innate and adaptive characteristics of the immune system, wherein a robot learns to detect vulnerable areas of a track and adapts to the required speed over such portions. A detailed study of the artificial immune metaphor is carried out and mapped onto the robot world. The robotics test bed comprised of two Lego robots deployed simultaneously on two predefined near concentric tracks with the outer robot capable of helping the inner one when it misaligns. The inner robot raises an SOS signal on misalignment. The outer robot aids the inner robot to regain it alignment exhibiting the innate immunity. The adaptive system within the inner robot learns to tackle the problem in future using Clonal Selection mechanism.