Conditional Q-learning algorithm for path-planning of a mobile robot

In classical Q-learning, the Q-table is updated after each state-transition of the agent. This is not always economic. This paper provides an alternative approach to Q-learning, where the Q-value of a grid is updated until a Boolean variable Lock associated with the cell is set. Thus the proposed algorithm saves unnecessary updating in the Q-table. Complexity analysis reveals that there is a significant saving in time- and space-complexity of the proposed algorithm with respect to the classical Q-learning.