https://github.com/smoothex/markov-decision-processes
A dungeon crawler game based on the stochastical model of Markov Decision Processes
https://github.com/smoothex/markov-decision-processes
game-development markov-decision-processes phaserjs typescript
Last synced: 4 months ago
JSON representation
A dungeon crawler game based on the stochastical model of Markov Decision Processes
- Host: GitHub
- URL: https://github.com/smoothex/markov-decision-processes
- Owner: Smoothex
- License: mit
- Created: 2022-06-14T11:37:30.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2022-08-02T09:34:30.000Z (almost 4 years ago)
- Last Synced: 2025-04-08T12:32:19.203Z (about 1 year ago)
- Topics: game-development, markov-decision-processes, phaserjs, typescript
- Language: TypeScript
- Homepage:
- Size: 37.6 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# This project has been forked from the [MTV](https://www.mtv.tu-berlin.de/menue/home/) GitLab server!
# You can play the game [here](https://pr.mtv.tu-berlin.de/2021sose-evakuierung)
# Short demo available [here](https://youtu.be/KSHAEq6LcC4)
### Game description
A little dungeon crawler game, inspired by a crowd evacuation simulator that was also based on the Markov Decision Process.
The fields represent different states that the player can switch to. The probability is implemented by a logic that splits the alien groups depending on which action field the player is standing on.
The action fields are all marked accordingly - there are split fields, group fields and death fields.
The goal is to reach the exit with as many aliens as possible and within the limit of movements.
### Theoretical background
The theoretical concept that this game is based on is called [Markov Decision Processes (MDP)](https://en.wikipedia.org/wiki/Markov_decision_process).
The core principle of the mathematical model is to either maximize the reward or the cost by switching states.
While in one particular state, the player can change the state by performing a certain action. The game is decision-making with the different outcomes (when changing a state) being partly random and partly controllable - this is what MDP is mainly used for.
### Credits
| Name | Role |
| ------ | ------------------- |
| Alexander Steffen | Product Owner |
| Christoph Rauchegger | Dev |
| Georgi Kotsev | MR Manager |
| Kevin Nguyen | MR Manager |
| Momchil Petrov | Dev |
| Phil Neujahr | Dev |
| Krutarth Parwal | Dev |
| Dennis Korolevych | Scrum Master |