Full-Stack Autonomous Robotic Vehicle

An educational hardware platform bridging analogue design, custom digital firmware, and mechanical test engineering.
Technical Project Management Analogue Design Firmware (C/C++) Test Engineering

During my tenure as an Assistant Lecturer for the Analogue Design module in 2022, I architected, designed, and managed a semester-long hardware project for undergraduate engineering students. The objective was to build a radio-powered, object-detecting, and object-avoiding remote-controlled robotic vehicle utilising a strict mixed-signal approach.

Operating as the technical project manager and lead full-stack architect, I was responsible for the end-to-end delivery of the project framework—from conceptualising the analogue/digital split and writing the core firmware, to managing the master Bill of Materials (BOM) and component sourcing, designing automated testing mounts, and providing technical mentorship to students with their own circuit debugging.

1. Mixed-Signal architecture & Full-Stack firmware

The core technical challenge was architecting a system where analogue circuitry and digital microcontrollers could operate harmoniously without ground loops or excessive noise coupling, seamlessly unified by custom firmware.

  • Analogue Control Logic: Half of the vehicle was driven entirely by analogue electronics. I designed sub-circuits for ultrasonic signal conditioning, analogue subtractors, and emitter follower circuits to directly drive the wheels based on sensor inputs.
  • Firmware & Digital Integration: Representing the digital half of the full-stack system, the analogue avoidance logic operated alongside an ESP32 microcontroller. I wrote the custom C/C++ firmware to manage telemetry, Bluetooth/WiFi communications, and precise digital pulse-width modulation (PWM) for wheel torque control.
  • RF Communications: I programmed and implemented the 433MHz RF communication protocols, establishing a reliable, low-latency link between the custom-built remote control interface and the main vehicle chassis.
  • Power Architecture: I implemented a unified power management strategy from a 6V battery source, designing undervoltage protection and isolated 5V voltage regulation to safely power both the sensitive analogue op-amps and the RF microcontrollers.
High-level system diagram demonstrating the isolated processing paths for the analogue object avoidance logic and the ESP32 digital remote control interface.
Remote-controlled car overview, with combination of analogue and digital design.

2. Simulation & Verification (LTspice)

Before physically prototyping the hardware, the analogue signal chain was mathematically modelled. I utilised LTspice to simulate the complete analogue control loop prior to physical implementation.

This phase involved modelling sensor outputs, verifying the analogue subtractor’s response to dynamic voltage changes, and simulating current sense logic. By front-loading the simulation, I ensured the baseline architecture provided to the students was mathematically sound and manufacturable.

Extract from LTspice schematic simulating the analogue control logic and signal conditioning stages prior to hardware prototyping, showing part of the design for the right wheel.

3. Automated test engineering & Mechanical design

I introduced automated testing methodologies to the module to assess the students’ final builds.

Instead of relying on subjective manual testing, I utilised Autodesk Inventor to design and 3D-print a custom mechanical testing mount to prevent the cars from moving during testing and standardise testing distances on the testbench. This allowed the robotic vehicles to be suspended and run at full throttle while live telemetry and analogue responses were measured objectively. Designing the assessment method alongside the mechanical fixtures ensured a repeatable, controlled environment for validating hardware performance under stress.

Simulated live testing environment showcasing the hardware running at full throttle while suspended on the custom 3D-printed assessment fixture for wheel rotation tests, while measurements and performance is monitored at three different test points with multimeters, and with an oscilloscope. The wheel slows down as an object comes closer, and comes to a full stop when the object gets too close.

4. Technical project management & Procurement

Beyond the technical design, a significant portion of my role involved acting as an engineering manager for the student cohort, ensuring smooth procurement and adherence to strict developmental milestones.

  • BOM & Supply Chain Management: I managed the master Bill of Materials, sourcing specific components ranging from MCP6242 op-amps and ESP32 dev kits to basic mechanical hardware. I navigated component availability to ensure the project remained viable within the academic budget.
  • Timeline Derivation: I broke the complex system down into a strict 12-week workflow. By establishing critical path milestones (e.g., physical current sense implementation in Week 2, DAC conversion in Week 4, and digital wheel control in Week 7), I ensured the project remained on schedule and realistic for the students’ academic schedules.
Excerpts from the formal project documentation showcasing the 12-week developmental roadmap and the comprehensive master BOM.