 The proposed strategy for shape assembly of robot swarms is based on the idea of mean shift exploration. When a robot is surrounded by neighboring robots and unoccupied locations, it will explore the highest density of nearby unoccupied locations in order to give up its current location. This is achieved through adaptation of the mean shift algorithm which is commonly used in machine learning for finding the maximum density of a given set of points. The proposed strategy has been tested with a swarm of 50 ground robots and compared to other strategies. It was found to be more efficient than existing strategies, particularly for larger swarms. Furthermore, the strategy can be adapted to generate interesting behaviours such as shape regeneration, cooperative cargo transportation, and complex environment exploration. This article was offered by Guli Bin Sun, Rui Zhu, Jaomar, and others.