horomanJumpSearch #1
Description
The goal is to implement an algorithm, that solves a shortest path problem on a two dimensional discrete map
where each node belongs to a category and has some straight and diagonal crossing costs assigned which are grater or equal the Euclidean distance. The categories are prioritized and the algorithm does not care about the costs of a category as long as the costs of the higher priority categories are minimized.
The algorithm is thought to be used on grid maps with areas of nodes of the same category and costs.
It will be built on the so called Jump Point Search which itself has it seeds in the label correcting algorithm, in particular on the A*-algorithm.
Here I would like to make a list, what needs to be done until the algorithm works. So this first entry will change over time.
- (done) Implement new heap
- (done) Implement new Node handling
- (done) exchange forced Neighbour check
- (done) Implement new cost calculation
- (done) Implement new findNeighbours
- verify and test