//Perform breadth-first search from initial state, using defined "is_goal_state" and "find_successors" functions

//Returns: null if no goal state found

//Returns: object with two members, "actions" and "states", where:

// actions: Sequence(Array) of action ids required to reach the goal state from the initial state

// states: Sequence(Array) of states that are moved through, ending with the reached goal state (and EXCLUDING the initial state)

// The actions and states arrays should both have the same length.

function astar_search(initial_state) {

let open = new FastPriorityQueue(function(a,b) { return a.estimated_total_cost < b.estimated_total_cost; });

let closed = new Set();

let fixed_step_cost = 1; //Assume action cost is constant

/***Your code for A* search here***/

/* Hint: A* is very similar to BFS, you should only need to make a few small modifications to your BFS code. You will need to add values to your augmented state for path cost and estimated total cost. I suggest you use the member name "estimated_total_cost" so that the above priority queue code will work. Call function calculate_heuristic(state) (provided for you) to calculate the heuristic value for you. See (included) FastPriorityQueue.js for priority queue usage example. */

//No solution found

return null;

}