This paper mainly focuses on the tracking and interaction strategy for an intelligent robotic rat. The target rat is tracked and positioned by using the binocular vision system of the intelligent robotic rat. An improved Camshift algorithm based on Kalman filtering is introduced to realize real-time tracking and positioning of the target rat. This paper designs a cognition system of the intelligent robotic rat based on SVM to achieve behavior recognition, behavioral classification and intelligent interaction of the target rat. Accordingly, this paper proposes an intelligent robotic rat tracking and interaction strategy based on the principle of bionics. The experimental results demonstrate that the tracking and interaction strategy proposed in this paper can realize the tracking and intelligent interaction between the intelligent robotic rat and the target rat, and thus confirming the feasibility and effectiveness of the strategy. The research provides biologists, neurologists and brain scientists with a research platform and assistant research tool.