← All projectsProblem statement Solution Architecture Key features Challenges Lessons learned
2025
AI-Enabled Smart Auto Injector
A syringe-based auto injector that adapts stroke length and speed to real-time depth sensing.
Manual injections risk inconsistent depth, stroke length, and speed, which can affect dosing accuracy and patient comfort, especially outside controlled clinical settings.
An automated, syringe-based auto injector that measures depth in real time and dynamically adapts its stroke length and injection speed, with live feedback on an OLED display. Built as a mini project at BMS College of Engineering (2025 to 2026).
An ESP32 microcontroller reads depth from an HC-SR04 ultrasonic sensor, then drives a stepper motor on a lead-screw mechanism through an A4988 stepper driver to regulate precise linear actuation. An OLED shield displays depth, stroke length, and speed while the firmware (C++) runs the control loop.
- Adaptive depth sensing with the HC-SR04 ultrasonic sensor
- Precision actuation via ESP32, A4988 driver, and a stepper + lead-screw setup
- Live OLED readout of depth, stroke length, and speed
- High accuracy, with an error margin under 0.5 mm versus manual injection
Converting noisy real-time depth readings into smooth, safe stepper motion, and tuning the lead-screw actuation so stroke length and speed stayed accurate to within half a millimetre.
Tight coupling between sensing and actuation is where embedded systems get hard. Calibrating the ultrasonic-to-motion pipeline early was the key to hitting sub-millimetre accuracy.