Projects | Engineering Horizons

Current projects · 2026–27

Find what you want to build

Each project below shows what the team is building, where it's at, and the skills you can pick up along the way. Open a project to see the full details.

New flagship project

Interactive Robot

An original interactive robot that can sense people and respond through motion, lights, sound, or other behaviors.

Mechatronics
Perception
Embedded
Controls
System architecture

Flagship

Main lead

Andreas

Overview

This is the project where we want to try something more original. The idea is an interactive robot (humanoid, creature-like, or somewhere in between) that can sense people and respond with motion, lights, sound, or other behaviors. The exact design is not locked yet. The team will decide what is realistic for Rev 1, then build around that instead of trying to cram every idea into the first version.

Photo coming soon

Concept images of other interactive robots. These are not the property of EHC.

Relevant skills & engineering concepts

System architecture: Requirements, subsystem definition, interfaces, tradeoffs, design reviews
Mechanical / Mechatronics: CAD, mechanisms, actuators, servo/motor selection, 3D printing, packaging
Electrical / Embedded: Power distribution, custom PCBs, microcontrollers, sensors, wiring, communication buses
Controls: Actuator control, motion sequencing, feedback, state machines, behavior control
Perception / Software: C++ / Python, OpenCV, cameras, tracking, gesture/object detection, ROS2 where useful
Human-robot interaction: Responsive behaviors, interaction design, safety, user testing
Integration: Bring-up, debugging, testing, documentation, subsystem coordination

Scope comes first

The first job is deciding what Rev 1 actually needs to do. The team will pick a realistic form and feature set, define the major subsystems, and build something we can actually finish and test this year.

Possible first-version features

A few expressive movements or gestures that work reliably.
Person, face, or object tracking using a camera or simpler sensors.
Reactions triggered by touch, proximity, or gestures.
Lights, sound, or display feedback that makes the robot feel responsive.

01Continued core project

4-Axis Robotic Arm

Mechatronics
PCB / power
Servo control
Embedded
Kinematics
ROS2 (later)

Continued

Project leads

Xander · Andreas · Geetha

02Continued core project

Sawppy Rover Upgrades

Improve a rover that already works: cleaner hardware, better sensing, and more autonomy.

System integration
PCB / wiring
CAD
ROS2
OpenCV
Autonomy

Continued

Project leads

Blake · Argi · Yuto

Overview

Sawppy is one of EHC's longer-running projects and already drives through a web interface. This year is less about rebuilding the rover and more about making it cleaner, easier to maintain, and more capable. That means fixing weak mechanical parts, cleaning up the wiring, improving the electronics, adding sensing/cameras, and starting to experiment with ROS2 and autonomy.

Photo coming soon

Relevant skills & engineering concepts

Mechanical: CAD redesign, structural improvements, mounts/enclosures, 3D printing, serviceability
Electrical: Power distribution, wiring architecture, connectors, soldering/crimping, custom PCB design
Embedded: Servo interfaces, Raspberry Pi / microcontrollers, sensor and camera integration
Software / Robotics: Python, ROS2, OpenCV, teleoperation, perception, autonomy
Systems: Retrofit design, interface management, reliability, testing, debugging an existing robot

Proposed upgrades

Clean up the electrical architecture and move toward a custom PCB / organized electronics bay.
Rework wiring and connectors so the rover is easier to troubleshoot, repair, and modify.
Redesign mechanical parts that are weak, awkward, or difficult to service.
Add a camera and experiment with OpenCV-based perception.
Start integrating ROS2 and build toward more autonomous behavior, navigation, or higher-level control.

Good fit if…

Pick Sawppy if you like improving something that already works. There is useful work here for mechanical, electrical, embedded, software, perception, and robotics members.

03Continued core project

Line Following Robot

Take the current prototype and turn it into a cleaner, more repeatable line-following robot.

ESP32
Sensors
Motor control
PID
PCB design

Continued

Main lead

Thomas

Sub-lead

Andreas

Overview

The line follower already moves, senses the line, and uses PID steering, but it is still very much a prototype. It uses an ESP32, a five-channel TCRT5000 IR sensor array, a TB6612FNG motor driver, and two geared DC motors. This quarter is about making it something we can run repeatedly without constantly babysitting it: better calibration, cleaner wiring, encoder feedback, stronger control tuning, and eventually a custom PCB.

Photo coming soon

Relevant skills & engineering concepts

Embedded: ESP32 / Arduino C++, GPIO, ADC/digital sensing, PWM, interrupts / timing
Controls: PID control, error calculation, tuning, differential drive, closed-loop behavior
Electrical: Motor drivers, power regulation, sensor wiring, breadboarding, PCB design
Sensors / Feedback: IR reflectance sensing, calibration, wheel encoders, noise / filtering
Mechanical: Mobile robot chassis, motor mounting, wheel/caster geometry, 3D printing
Testing: Serial debugging, repeatable track tests, parameter tuning, performance comparison

Proposed upgrades

Keep tuning the PID controller and improve sensor calibration so the robot tracks the line more consistently.
Add wheel-encoder feedback so the two motors behave more consistently.
Improve or redesign the sensor array if sensing becomes the main limitation.
Move off the breadboard toward a custom PCB and cleaner power / connector layout.
Refine the chassis and packaging so the robot feels like a finished platform instead of a prototype.

Good fit if…

Pick LFR if you want embedded programming, sensors, motors, PID control, PCB work, and lots of hands-on testing. It is a good controls project because you can see the effect of a code or tuning change almost immediately.

04New core project

SLAM Robot

Build a rover that can map a room, determine its location, and navigate autonomously.

LiDAR
Odometry
Sensors
Navigation
Embedded controls

New

Main lead

Palak

Sub-lead

Geetha

Overview

This is a new project, so the team gets to make a lot of the early design decisions. We want to build a differential-drive rover that can be driven manually, estimate its own motion, build an indoor map with 2D LiDAR, and then navigate around that map on its own. The challenge will be getting the mechanical, electrical, embedded, and software pieces to agree with each other.

The progression for this project is still being planned out. A key milestone this quarter is setting up the architecture, understanding the underlying system and project requirements, then building a proof-of-concept physical design using already manufactured components.

Photo coming soon

Concept images of other open-source SLAM robots.

Relevant skills & engineering concepts

Robotics software: ROS2 (TBD), nodes/topics, TF frames, RViz, SLAM Toolbox, Nav2
Localization: Odometry, coordinate frames, sensor fusion, uncertainty, mapping/localization
Sensors: 2D LiDAR, IMU, wheel encoders, calibration, sampling / noise
Embedded / Controls: ESP32 or STM32, encoder reading, PID wheel-speed control, motor drivers
Electrical: Battery / regulation, power distribution, grounding, wiring, integration
Mechanical: Differential-drive chassis, sensor mounting, wheel alignment, packaging

Potential stretch work

If the core robot is working early, add a camera for object detection or semantic mapping, try another localization sensor, or use the platform as a future testbed for controls, computer vision, and machine learning.

Found a project you like? Let us know on the interest form.

Member interest form