Infinite Terrain Generator

A Three.js experiment in dynamically re-rendering terrain as the camera moves throughout the universe.

Conway's Game of Life

A fully configurable simulation of the Game of Life created using JavaScript and rendered using HTML5's canvas element.

Planet Generator

A random planet is generated using Perlin noise applied to the surface of a sphere. Each terrain patch in the mesh is colored using a height map to create a simple but natural looking planetary surface. This project uses the Three.js graphics library to create this 3D scene.

Gravity Simulator

A simple gravity simulator that demonstrates the motion of three bodies. This project uses the Three.js graphics library to create this 3D scene.

Frisbee Simulator

This Frisbee Simulator accurately models the flight of a frisbee through the air. Optimization techniques are used to solve for an initial set of throwing parameters to aim a frisbee throw to a target somewhere in 3D space.

Inverse Kinematics Using Optimization

Through the use of function optimization in MATLAB, the position and orientation of a 3-dimensional snake robot is controlled in order to reach into an environment and grasp an object. The snake robot consists of a serial chain of kinematic links, of arbitrary number and length, where each joint is controlled with a set of Euler angles (roll, pitch, yaw). The environment is filled with spherical obstacles that the robot must avoid in reaching the obstacle.

Inverted Pendulum Control

This project regards dynamics, control, and robot learning techniques to both model and control a cart pole (inverted pendulum) system through the use of a simulated dynamics engine. The cart has a mass of 1kg and a horizontal rocket thruster to move in the horizontal x-direction. A pole is attached to the cart with a joint. It has a mass of 0.1kg and is 1m long (and is infinitely thin).

3D State Estimation and Control

This project deals with tracking/representing the 3D orientation and dynamics of an “alien artifact” tumbling through space. The goal is to use markers on the corners of the artifact to estimate its dynamics and future trajectory. Finally, we are to control a spacecraft to land just above the surface of the artifact as it tumbles through space.

Extended Kalman Filtering

This project is an extension of "3D State Estimation and Control" as it deals with estimating the state of an “alien artifact” tumbling through space. However, here we focus on state estimation in the presence of sensor noise and occlusion. The goal is to use an Extended Kalman filter to produce optimal estimates of the position, orientation, and velocity of the artifact using noisy data of the markers on the corners of the artifact.

Software Projects