D* Lite Algorithm

Summary

For the Intelligent robotics module I implemented the D* Lite Algorithm in python from this paper. D* Lite offers an advantage over A* because you don't need to rerun the entire path if the path is blocked. This was useful in the project group I was working with.

Repository link, but private because of university policy.

Result

The video below shows how it finds a path and recalculates when there is an obstacle. I have uploaded this from another user than the other videos. g is the cost from the point to goal, whilst rhs is one step ahead, which is why it is a bit wider than g. Updates is incremented everytime there is an update to either value. The map where this was explored is shaped as an I, or sideways H.