wattik / plannig-hw

Planning in AI

This is a collection of source codes implementing a solution to homework tasks given in the course Planning in AI /PUI/ at CTU. All code has been unless otherwise stated written by myself. Please, do not copy the solution.

Contents

Source Folder	Task Name	Approach	Description
ass1	STRIPS Planning Solver	LM-Cut	The task is to implement a solver of STRIPS planning tasks based on the state-of-the-art heuristic LM-Cut.
ass3/src/wumpus/agent/	Planning in Stochastic Environment (MDP)	Monte Carlo Tree Search (UTB)	In a stochastic environment, an agent plans its actions taking into account: randomness in its actions and probabilistic behaviour of adversarial players. In this task a robot is asked to find a policy that helps him to gather as much gold as possible while navigating itself in a grid-world with monsters and dangerous pits. Also, the result of the robot's actions is non-deterministic.