RRT Tree Growth Animation

The initial tree is a single node (red dot). We grow the tree by selecting a random target point q, finding the nearest node n to that target, and extending the tree from n by a short distance towards q, creating a new node n'. As this process repeats, the tree rapidly explores the entire space. This shows why RRTs are an efficient search technique. With enough time the tree will eventually fill the entire space.

Animation done by Steven LaValle of the University of Illinois at Urbana-Champaign.