Xyon Labs

Work

Robots, perception systems and learning pipelines we have designed and built. Hardware projects and simulation projects are tagged.

Voyager: Mobile robot platform built from the ground up

Mobile robotics Hardware

Voyager

A differential-drive research platform: chassis, electronics, firmware and ROS 2 autonomy, all designed in-house.
Hydra: Six-axis manipulator from CAD and PCB to real-time control

Manipulation Hardware

Hydra

A 6-DOF robot arm designed, printed, wired and programmed in-house, with MoveIt 2 planning and MoveIt Servo real-time control.
Multi-floor receptionist robot: A semi-humanoid that guides visitors across a whole building

Mobile robotics, HRI Hardware

Multi-floor receptionist robot

A semi-humanoid receptionist that greets visitors, navigates crowds and rides elevators between floors on its own.
Humanoid control: Unitree G1 whole-body control with GR00T and SONIC

Humanoid robotics Simulation

Humanoid control

Loco-manipulation on the Unitree G1: a vision-language-action model as the brain, a whole-body controller as the body, developed simulation-first.
LiDAR perception: Detecting and tracking people directly from laser scans

Perception, AI Hardware

LiDAR perception

An end-to-end AI system that detects, classifies and tracks people and objects from LiDAR data in real time, deployed on target hardware.
Autonomous surface vessel: BlueBoat USV with real ArduPilot, model-predictive control and learned perception

Marine robotics, control Simulation

Autonomous surface vessel

A Blue Robotics BlueBoat crossing open water through a buoy field, driven by real ArduPilot SITL, an MPPI controller and a CNN that sees only the forward camera.
Soft-material handling: Dual Kinova Gen3 arms learning from demonstrations

Manipulation, imitation learning Simulation

Soft-material handling

Two Kinova Gen3 arms learning to handle soft, deformable material from teleoperated demonstrations in simulation.
Dual-arm manipulation research: A bimanual platform for reinforcement learning on contact-rich tasks

Manipulation, reinforcement learning Simulation

Dual-arm manipulation research

A two-arm simulation platform for training reinforcement learning policies on two-handed tasks, with stacking and sorting as benchmark tasks.
Language-conditioned manipulation: SmolVLA pick-and-place on dual Franka arms

Robot learning, VLA Simulation

Language-conditioned manipulation

A vision-language-action pipeline in which two Franka arms follow typed instructions to pick, place and sort objects.
Competition robotics: Sri Lankan Robotics Challenge and RoboRoarZ

Competitions Competition

Competition robotics

Custom robots for the Sri Lankan Robotics Challenge at school and university level, and reconfigurable-robot work at RoboRoarZ Singapore.
Early work: Vision-guided mobile robots and a tidal-energy concept

Mobile robotics, concepts Simulation

Early work

An ESP32-CAM robot navigated by a vision server, a Webots simulation, and the AlgaGen hydrogen-from-tides concept.

More projects

Client and research work without published footage.