Response 3

To better understand planning, we're going to watch a talk by Jeff Orkin from GDC 2005. This is the talk that introduced Goal Oriented Action Planning (GOAP) to the game community. It's really an adaptation of the planning paradigm that had been developed in academia for decades. But we're going to look at the GOAP talk because it introduces planning specifically in the context of games (though in the context of first person shooters, which I recognize are not everyone's cup of tea), and because one of the goals of the Unity Planner is to make an advanced GOAP-like facility available to Unity developers. In future classes we'll backtrack to briefly review search more generally (so that we can see that planning is just a specialization of searching a state-space), and we'll look more deeply into Monte Carlo Tree Search (MCTS), which is a mainstay of modern Game AI and which the Unity Planner builds on and generalizes.

The GOAP talk is on the GDC Vault Links to an external site.. Unfortunately, the video of the talk no longer exists, so there is just the audio. However, Jeff Orkin has made his slides from the talk available Links to an external site.. This is a zip file containing his slides and the videos from his talk. So you can follow along with the audio, looking at the slides and playing videos as appropriate. After watching the talk, give the PDF you'll find in the same folder a skim read. It covers the same material as the talk, but you may find that reading one or more parts of the paper in more detail will aid your understanding of the material. 

Submit a 1 to 1.5 page PDF answering the following questions. Include a header in your document for each question. You can work on this in pairs if you wish to facilitate conversation, but if you do, you should both submit the reading response so that I have it in canvas for each of you. Additionally, be sure to include both your names at the top of your response.  

Questions

• Briefly summarize the talk/article.
• What was most convincing or helpful to you? (Where did you see or learn something new with the authors help? Which examples or evidence was most effective?)
• What was *not* convincing? (Where did the authors overlook something? Where would you challenge them?)
• What questions about planning do you still have? (Could be about defining planning states, or defining actions including preconditions and effects, or about how plan goals work, or how the search works, etc.). 
• Any other questions/comments?