This example demonstrates a partially observable scenario using a grid world. The red circle represents an agent that is trying to reach the yellow star at the top of the grid. Each step incurs a cost to the agent, but only a small region around the agent is observable.
Click on a grid cell to move the agent. The numbers represent the cost of moving into that cell. The total cost is displayed at the bottom.
Imagine how the agent might choose to make decisions based on the information it observes and what it believes to be the remaining cost to the goal. How likely is a road block along each route? Is it bad enough to turn around? How might different agents make different decisions?