1 – Example Route Finding

Now let’s see how the definition of a problem maps on to the route founding domain. First, the initial state we’re given, let’s say, we start off in Arad. And the goal test, let’s say, that the state of being in Bucharest is the only state that counts as a goal. And all other states are not goals. Now the set of all the states here is known as the state space. And we navigate the state space by applying actions. Actions are specific to each city, so when we’re in Arad, there are three possible actions to follow this road, this one, or this one. And as we follow then, we build paths or sequences of actions. So, just being an Arad is the path of length 0. And now, we can start exploring into space and add in this path or length 1, this path of length 1, and this path of length 1. We can add in another path here of length 2, another path here of length 2. Here’s another path of length 2, here’s a path of length 3, another path of length 2 and so on. Now, at every point we want to separate the state out into three parts. First, the ends of the paths, the farthest path that been, have been explored we call the frontier. And so the frontier, in this case, consists of these states that are the farthest out we’ve been explored. And then, to the left of that in this diagram, we have the Explored part of the state. And then, off to the right, we have the Unexplored. So let’s write down those three components. We have the Frontier, we have the Unexplored region, and we have the Explored region. One more thing, in this diagram we’ve labelled the Step Cost of each action along the route. So, the Step Cost of going between Neamt and Lasi would be 87 corresponding to a distance of 87 kilometers. And then the Path Cost, is just the sum of the Step Cost. So the cost of the path of going from Arad to Oradea would be 71 + 75. [BLANK_AUDIO]

Dr. Serendipity에서 더 알아보기

지금 구독하여 계속 읽고 전체 아카이브에 액세스하세요.

Continue reading