CSCI 5611 Project 1

Agents Collision in 3D Enviroment

Video

Uploaded in this repository, Proj1.mp4

List of Features

Part 2

Single Agent Navigation
3D Rendering and Camera
Improved Agent & Scene Rendering
Orientation Smoothing
Multiple Agents Planning+

Part 3

Crowd Simulation

Others

Loading Screen

Art Contest

User Guide

Press 'r' to reset the scene

Press 'm' to change the render mode.

mode 0 - only render the direction from agent current location to goal
mode 1 - add the path of agent
mode 2 - add the collider of obstacles and agents

Arrow Keys can rotate the camera along x axis (Left and Right) and y axis (Up and down)

Press 'w' and 's' move camera front and back

Press 'a' and 'd' move camera left and right

Press 'q' and 'e' move camera top and down

Press '=' to add agent (maximum number for agent on the scene is 20)

Press '-' to take away agent from scene

Discription of Features and Timestamp

Single Agent Navigation

Timestamp: 00:00-01:05

Agent (i.e. tank) is implimented with path finding and it can short-cut its path while it is moving to the goal.

Once the agent reachs its goal, it will altomatically generate a new goal and navigate to it.

It also has the orientation smoothing function, which allows it to turn smoothly. (Angular velocity is 180 degree per second)

3D Rendering & Camera

Timestamp: 01:05-02:05

Camera can rotate along its x axis and y axis and it can translate based on its x axis, y axis and z axis.

Models and Textures - Improved Agent & Scene Rendering

windmills - stands for the obstacles with larger radius

silos - stands for the obstacles with smaller radius

tanks - stands for the agents

Models come from https://quaternius.com/index.html

Desert-Bush - floor texture

Texture comes from https://assetstore.unity.com/packages/2d/textures-materials/glovergames-free-ground-textures-135418

Multi Agent Planning and Crowd Simulation

Timestamp: 2:05-4:05

Simulation is generated on the random generated scene. Press '=' to add agents and press '-' to take agents away. Whenever a agent(i.e. tank) reaches its current goal, it will altomatically generate a new goal on the scene and navigate itself to the new goal.

Difficulties Encontered

Shader program is hard to impliment in this project, so I finally gave up.
The insertion point (i.e. cursor) of the editor is not in its position for Processing 4.0b1. I will try the lower vesion in the next project.

wenjie-zhang08 / csci-5611-project-1 Goto Github PK