Final System Integration and Robot Design

---

Photoshoot

---

Robot - Hardware Design

• Bill of Materials for Main Components
• 1 × Arduino Uno
• 2 × Continuous Rotation Servos
• 1 × Battery Pack
• 1 × Pushbutton
• Positioned as low as possible to give a low center of mass in z-direction for stability
• 3 × Digital Line Sensors
• 3 × Long Range IR Sensors
• 1 × IR Hat
• 1 × IR Sensor
• 1 × Microphone
• 1 × Radio transceiver
• 1 × Camera (*)
• 1 × FPGA (*)
• 1 × Permanent Switch (*)
• Budget

• Fail Safe Switch
• A simple push-button switch is wired into a digital I/O pin
• When pressed, this overrides 660 Hz detection and advances the robot into the main loop
• Althogh we have never had any issues with our microphone detection circuit, this is a valuable fail-safe system
• (*) Camera and FPGA
• We made the tough decision to not mount these components for the competition
• Although they are fully integrated into our software, our Treasure Detection algorithm does not work due to a broken camera
• Since there is a penalty for false detection, we decided to prioritize other tasks and not mount these components
• If our Treasure Detection had worked, we would've mounted the components as follows:
• Camera:
• The camera would've been mounted on the right-hand side next to our breadboard, since our DFS algorithm prioritizes left over right
• The SCL pin, A5, is also used by our microphone detection circuit
• To allow for this, we came up with an innovative solution with a permanent switch:
• The camera is only programmed once, immediately upon power-on
• Once this is done, the SCL pin can be switched to an input-pin and used for 660 Hz detection
• Using the manual switch, we would (1) turn the robot on, (2) wait 20 seconds for the camera to be programmed, (3) flip the switch to connect the microphone circuitry
• This method worked flawlessly during our testing
• FPGA:
• The FPGA would've been mounted next the the Arduino on the right-side, with the Arduino moving left to create space
• This would allow for easy connections between the FPGA and the Camera
• Wiring
• Proper wire types
• Solid Core = for breadboard connections
• Stranded Core = for flexible component connections
• Wiring Scheme

• We made every effort to keep wires as short as possible
• Pinout

Robot - Software Processes

---

Base Station

The purpose of the Base Station is to display the location of the robot, and the information it collects, as it traverses through the maze. This is recorded on a GUI, which is being run on a computer plugged in over USB.

• Hardware:
• Arduino Uno
• Nordic nRF24L01+ transceiver
• Computer running GUI script
• Software:
• Initialize radio and serial connections
• Contains a Maze object, which represents the physical maze
• This is identical to the Maze object on the robot
• Once the robot collects information to update its on-board Maze, it transmits the requisite message to update the Base Station's Maze object
• Calls function maze.getGUIMessage() to update the GUI
• Recently, we added functionality to display when other robots are detected
• Pseudocode:

          void setup(){
              initialize radio
              initialize serial
          }

          void loop(){
              while !radio.available()
                  wait for radio transmission
              receive transmission
              decode instruction
              update internal Maze object
              determine GUI update message and print to serial
          }
        




---




Future Improvements (aka Wish List)

• Printed Circuit Board (PCB)
• Tyler had started work on an Arduino Shield for our robot in Altium Designer
• The goal was to use headers and terminal blocks to eliminate (or reduce) the number of breadboard connections
• The board shape was outlined, components imported, and schematic created
• Due to time constraints and wire changings, however, we were unable to complete this

PCB schematic

PCB footprint

3D model of PCB with components
• Structural Components
• Brian and Kenneth had begun designing new structural and mounting hardware to be 3D printed
• This would've eliminated the issues we were having with our ball-bearing, as well as with attaching sensors
• Unfortunately, other priorities got in the way and we were unable to finish this


Orbit View



Exploded View