Don’t Teach Me Sudoku

I’ve been playing the New York Times Sudoku on occasion, and it takes me almost an hour to solve the medium puzzle. Fun isn’t fun until its ruined, so as is de rigueur, I started kicking around the idea of getting a computer to solve sudoku puzzles. Of course sudoku-solvers already exist on the internet,…

Dueling Cargo Bikes

Introduction In my last post I looked at the practicality of a human powered cargo bike. The short of it is that it’ll work okay if you live where its flat, and you don’t mind going slow. If you want to feel the wind in your hair and climb hills with ease, you’ll want to…

Electric Cargo Bikes & Their Undeniable Allure

Introduction While biking in the city, I am frequently passed by electric and non-electric bicyclists. I glare at the offending passers-by for confirmation of tell-tale signs of an e-bike. Any large dark barnacles protruding from the frame will set me at ease. A motor hanging from the bottom bracket, a battery bolted to a downtube…

Interpreting a Data Driven Flight Dynamics Model

Introduction In my last post I developed a flight dynamics model (FDM) using data from a fixed wing RC plane containing the Pixhawk1 running Ardupilot firmware. I suggested that I should look into the following: Visualize a learned model’s relationship between control inputs and state variables. For instance, does the model learn what we might…

Learning a Flight Dynamics Model

Introduction & Background What is a flight dynamics model? Flight dynamics is the science of air vehicle orientation and control in three dimensions. The three critical flight dynamics parameters are the angles of rotation in three dimensions about the vehicle’s center of gravity (cg), known as pitch, roll and yaw. Wikipedia – obviously A good…

Precise Locating on Complex Geometries

Sometimes you might want to put something on a specific spot on a complex shape. Like a hole for passing wires, or an instrument for measuring load, strain, acceleration, or some other phenomenon. The Problem How do I easily put a mark on some real life complex surface and know its CAD coordinates and vice…

General Aviation Hybrid Electric Feasibility

A recent adoption of electric vehicles by the US automotive industry indicates that electrification could become viable in general aviation. This is an exciting development for an industry that has not kept pace with its automotive, or commercial aviation counterparts. What about Fully Electric? Battery energy density (measured in watt hours per kilogram) is too…

Physical Simulation Optimization

In my last post, I applied a reinforcement learning algorithm called Q-Learning to maximize velocity in simulations attempting a human powered land speed record. This algorithm was limited by: An incomplete physics model. A sub-optimal reward function. A single driver limit. Improving the algorithm and simulation in these three ways is the focus of this…

Reinforcement Learning for Speed

I have recently written a paper on applying Q-Learning – a reinforcement learning algorithm – to the acceleration problem brought up in my Going Fast on Your Own Power series. The paper’s abstract, the paper itself, and a corresponding presentation are below. Abstract In this report, Q-Learning is used to maximize velocity and minimize driver…

Going Fast on Your Own Power – Design

In the last two posts I went over what it takes to go fast on your own power. I found that its possible to maintain world-record breaking speeds for a fit cyclist in the right vehicle, but accelerating to those speeds is nearly impossible in current vehicles unless you are a top level cyclist. In…