In this paper, we present a comparative study of three distributed strategies for task allocation in a multirobot system. The objective is to determine the course of action for each robot and the targets it needs to service. A theoretical section is provided to support the dynamics of these techniques and some of the results. The first technique is a neural network-based approach, known as self-organizing map (SOM), that assigns targets to the robot on the basis of competition. The second technique is a combinatorial technique, known as the Hungarian method for solving assignment problems. The third technique is an integer linear programming-based optimization approach that tries to minimize the cost of task allocation. By implementing these three techniques, we observed that SOM tends to yield better results in terms of cost of assignment and execution time, but suffers from lack of fairness and workload balancing. In contrast, the other two methods fulfill the two criteria, but at the expense of a relatively higher cost.