Taking place in ancient Egypt, a peaceful society of cat people has been invaded by roman rats led by the wicked tyrant, Julius Cheesar. In order to rescue baby kittens that have been taken purrisoner, two courageous cats must sneak past rat guards and solve the pyramid’s puzzles as they go deeper…into the catacombs.
Building a behavior tree system from scratch was a challenge but enabled a quick turnover for very basic stealth AI involving a complex array of layered states and player sensing.
Rapid development and iteration of player abilities and gameplay feature made for very rewarding testing cycles to find the core of our game.
In managing project direction and working closely with as much of my team as possible I helped ensure a coherent team vision that included as many perspectives as possible.
Aiming for ambition, the team concluded our conception phase with a cooperative stealth puzzler. We would need level design, play design, environment art, 2 characters, stealth mechanics and ai, and networking.
While I hardcoded these trees without a GUI, I learned my lesson from the previous project and made behavior trees. This would be an important decision for making the guard AI come together in short time, taking only a week or two to prototype.
While we planned for potentially complicated elements, the basics of a stealth AI is really quite simple. Using a blackboard and updating it with stimuli information when necessary made this a breeze. Taking detection information from the blackboard and accumulating that into a detection value then made the core of the game a matter of memory and thresholds.
The most difficult part was the system by which the AI would remember the last known location of the player. A series of discussions on who the AI should pursue and how much information it should keep helped to define a system so that guards would round the last corner they saw you dart behind but would handle a number of edge cases in ways that minimized player frustration.
During this project, we had a number of challenges we had to deal with. Cutting scope hard helped ease pressure on designers and freed more resources to ensure networking stability. It also ensured the art team was always working on material that was important for the experience
Helping to lead meetings and discussing direction and needs with team members was essential for ensuring a cohesive vision such that, even though I could not be present for the final week and the greenlight of our project, we met our final deadlines and pitched the vision accurately and completely.