P17241: Autonomous People Mover IV
/public/

Build & Test Prep

Table of Contents

< back to home page

Updated Customer Requirements

Customer Requirements

Customer Requirements

Click here to see full updated file.

Updated Engineering Requirements

Engineering Requirements

Engineering Requirements

Click here to see full updated file.

Updated Test Plans

ER Test Plans

ER Test Plans

Click here to see full updated test plan.

Risk and Problem Tracking

Issues

Issues

Click here to see full updated document.

Updated Fuse and Wiring

Front Wiring Diagram

Front Wiring Diagram

Click here to see updated document.

Mounting plate for the fuse block and electrical components in the front.

Front Electrical Mounting Plate

Front Electrical Mounting Plate

Click here to see updated document.

Brake Frame Drawing

Brake Frame Drawing

Brake Frame Drawing

Click here to see full updated document.

Software Planning

Updated ROS software layer flow chart to include the new additions of encoders and break, steering, and throttle user inputs.

ROS Software Layer

ROS Software Layer

Click here to see full updated document.

New software flowchart for additions to the Arduino code to accommodate the encoders and braking sensor.

ROS Software Layer

ROS Software Layer

Click here to see full updated document.

Wheel Encoders

Encoder Ring

Multiple prototypes were constructed for the encoder ring to ensure the proper operation of the final design. The last of which was the most functional. 20 magnets were purchased for testing and the following tests performed.
Prototype Wood Encoder Ring

Prototype Wood Encoder Ring

A pitch test was performed using 20 magnets at the intended operating pitch in the wood prototype to ensure that the sensor was capable of detecting the magnets at speed. The cart was jacked up to lift one rear wheel and the throttle pressed to the floor to test double the theoretical maximum operating speed.

A Wheel run-out test was performed using 20 magnets inserted evenly around the wheel to ensure that the sensor was capable of detecting across the entire rim of the wheel. The prototype was slightly too large and therefor the ring did not set into the wheel fully. Even with this disadvantage the sensor was not skipping any pulses.

Results showed good detection of the magnets on the rim but the distance between the sensor and magnet faces was on the order of a millimeter and was very close to touching. To ensure the proper operation of the encoders in the final design, the strength of the magnets and the sensitivity of the sensors were increased.

Final construction used a PVC sheet machined on the CNC router at the Construct.

Encoder Ring Drawing

Encoder Ring Drawing

Click here to see full document.
Final Encoder Ring

Final Encoder Ring

This encoder ring will snap fit to the inside of the rim during testing and can be permanently attached with epoxy when testing is finished. The ring has all 200 magnets press fit in opposite orientations.

Sensor Bracket

An acrylic prototype of the sensor bracket was laser cut and bent at the Construct. The purpose of the prototype was to ensure that the critical dimensions of the bracket were correct. This was not the case and a Dremel was used to tune in the bracket pending the final construction of the bracket out of aluminum.
Sensor bracket with sensors

Sensor bracket with sensors

Encoder Sensors

Sensor testing was executed using a 0.1in headers some wire and a bench power supply. A set of helping hands was used to hold the sensor in the correct spot.
Sensor Testing

Sensor Testing

Final construction was accomplished by first mounting the DRV5013 sensors in a 3D printed plug. The purpose of the plug is to orient the sensor in the middle of the body tube so the phase angle between sensors at 90 degrees. A decoupling capacitor was soldered to the front of the sensor such that there is little chance of shorts to the chassis and as close to the sensor as possible. Cabling was soldered to the leads of the sensor and heat shrink was used to insulate the connections. The body tube can be potted with epoxy once operation of each sensor and testing is verified.

Sensor Detail View

Sensor Detail View

Encoder Electrical

Initial revision hardware was built on a proto-shield. This can be mated to the existing PCB through the debugging headers. The hardware used includes the LS7366R, this chip is a 32 bit quadrature encoder counter with an SPI interface. There is one for each wheel. The chip was used in order to hasten the development time, and reduce the risk of errors in counting the pulses. The chip has a 40 MHz oscillator as a filter clock. There are also 200 ohm pull up resistors for each of the 4 sensors. The drive current is intentionally high, 15mA in order to prevent noise issues. The DRV5013 in general has 30mA capable open collector drive.
Encoder Electrical

Encoder Electrical

Encoder Software

Initial software testing went extremely well, the code is able to read the encoder counts on the both the left and right channels. The sensor wires were temporarily swapped from the left to the right channel pending the construction of the final sensor hardware.
Encoder Code

Encoder Code

Design Review Materials

Include links to:

Procurement Plan

Procurement Plan

Procurement Plan

Click here to see full document.

Schedule

Phase Deliverables:

1. Build Prep:

2. Subsystem Functional Demo

3. Preliminary Integrated System Demo

4. Full System Demo

5. Customer Hand-off

Click here to see Project Schedule

Closure Plan

Closure Plan

Closure Plan

Click here to see full document.

Bill of Materials

Click here to see full document.


Home | Planning & Execution | Imagine RIT

Problem Definition | Systems Design | Preliminary Detailed Design | Detailed Design

Build & Test Prep | Subsystem Build & Test | Integrated System Build & Test | Integrated System Build & Test with Customer Demo | Customer Handoff & Final Project Documentation