Meida*

#REDIRECT IDA

Memory extended IDA*

In situations where the main memory is too small to hold all the nodes reached by IDA* MEIDA* can be used. IDA is International Data Address

This algorithm runs in two steps:

First step: Fill priority queue D with at most m nodes running IDA*

Second step: expand the minimum from D and insert its successors into D if - D is not full - and f value of successor ≤ maximal f value in d continue until D becomes empty or goal found

The last expanded nodes f value is the threshold for the next IDA* run.

Complexity:Expands at most O(n+m+n²/m) nodes