Quadrotor
From Null Space Labs
Contents |
Design
- 4 Motors ESC
- ARM Cortex-M3 80Mhz
- Sensors: Gyro, Accelerometer, Compass, Temperature, GPS & Sonar(to come in later design)
- Xbee 802.15.4 Wireless Communication
Plan
Project is current in Phase B-9.
Design Diagram: File:Quadrotor uav arko.vsd
Phase A - Hardware Development
- Build a one motor/ESC setup, vary speed (Prototype with Arduino)
- Build a two motor/ESC setup, vary speed (Prototype with Arduino)
- Design board with new TI chip, sensors, power management, etc.
- Layout, mill and test proto board
- Build a two motor/ESC setup teeter-tot (Pitch) and wire up to proto board, test all hardware
Phase B - Modeling & Software Development
- Create library (or use Maple IDE)
- Create Kalman filter for all sensor inputs
- Create basic flight control/stabilization code
- Balance one degree of freedom very well (Pitch)
- Code Clean Up
- Build a rig to support two degrees of freedom (Pitch and Roll)
- Balance two degrees of freedom (Pitch and Roll) very well
- Modify rig to support three degrees of freedom (Pitch, Roll, Yaw)
- Balance and stabilize all three axis, add fourth degree of freedom (Altitude)
- Create wireless communication with computer/controller and command set
Phase C - Testing/Revising
- Test and improve board layout, filtering, design
- Design/CAD new chassis, mount and update model
- Add any extra features
- Test, Test, Test
Phase D - Package and Document
- Document all parts, assembly notes, and code
- Create cost summary and links to buy parts
Phase E - Complex Flight
- Follow flight path defined by user
- Do tricks?
- Stage 6 to CDM, separation at level 4
Phase F - ???
Phase G - Profit
Chassis and Hardware
- Build teetertot 1 degree of freedom (pitch)
- Picture and build instructions coming soon..
- Build 2 degrees of freedom balance (pitch and roll)
- Build 3 degrees of freedom rig (pitch, roll, yaw)
- Build 6 degrees of freedom rig (hang on a string?, pitch, roll, yaw, long, lat, alt)
Board Layout and Schematic
- Board layout and schematic to be posted soon..
- List of Parts to be posted soon..
State Estimation & Kalman Filter
To obtain a clean state estimation one must first filter the sensory inputs then produce a state model based on that.
- Create Kalman filter
- Here's a simulation of a gyro and accelerometer filtered to give a clean angle estimation: File:Kalman gyro acc.m
- Create state model
Control System
Hardware links
- Processor: TI LM3S9B96 - ARM Cortex-M3 80Mhz
- Gyro: L3G4200D
- Accelerometer: ADXL335
- Compass: HMC5843
- Temperature: TMP35
- Crystals: 16 Mhz Crystal
- Crystals: 25 Mhz Crystal
- Power Regulator Buck: TPS54231
- Level Logic: BSS138
- Wireless: Zigbee Module
Software Links
- Maple IDE: Maple IDE
- QGroundControl: qgroundcontrol
- IAR: IAR ARM
Pictures & Videos
First Stable Flight with APM!
New Chassis!
First flight test with APM
New Frame!
Fixed Kalman Filter Lag issues! Small setback, but the control should be much cleaner.
New Mount! Finally Rotational X,Y,Z!
New Chassis!! Finally have 4 rotors!
New board in action!
<---- New boards!!
Balancing with a Kalman Filter and PID loop tuned!
Balancing for the first time! (no Kalman Filter)
<---- Time to get balancing!
<---- Grrrrrr!
Powered up the motors and able to read accelerometer!
First time firing up the board and the code uploaded just fine! Test code works!
<--- Nsl Drone Board Rev.2 assembled now with a 1.2v reg and through-hole plating!
Beginning code works! Able to read multiple ADC inputs and output multiple PWM outputs!
<--- Milled Board and assembled!
<--- Dev Board in drone simulation mode
.JPG)
<---- Oh yeah!
<---- Almost there!
<---- :)
<---- CAD of the concept
Achievements Unlocked
