AStarGoapPlanner

Implements a Goal-Oriented Action Planning system using the A* algorithm. A* works by finding the optimal sequence of actions to transform an initial state into a goal state while minimizing total cost. See https://en.wikipedia.org/wiki/A*_search_algorithm

The algorithm works as follows:

1. Start with the initial world state

2. Maintain an open list (priority queue) of states to explore, prioritized by f-score

3. For each state, explore all achievable actions, calculating:

• g-score: The cost accumulated so far to reach this state

• h-score: A heuristic estimate of the remaining cost to reach the goal

• f-score: g-score + h-score (total estimated cost)

4. Always expand the state with the lowest f-score first

5. Track visited states and their best known costs to avoid cycles and redundant exploration

6. Continue until finding a state that satisfies the goal conditions

The implementation ensures finding the optimal (lowest cost) sequence of actions by properly tracking path costs and using an admissible heuristic function.