Aalborg University Kugle

Not for sale

Not for sale (research)Via SSH / web / bridgeOpen-source ballbotROSMPCResearch

Not for saleShipping: N/A.

Remote control: The published ROS driver runs on the robot's onboard computer. A laptop joins the ROS network over Wi-Fi to send velocity and path commands (only relevant to anyone who rebuilds one).

LEARN MORE (Kugle-MATLAB, GitHub) ↗

Overview

A ballbot built as a human guide robot for a 2019 Aalborg University master's thesis, notable for a quaternion-based model, sliding-mode balance control and an MPC path follower. It uses two lidars to localize and detect people. The whole stack (models, firmware, ROS and Gazebo) is open source, which makes it the best public reference for anyone building a ballbot.

Type: Rolling robot · Research & concepts
Maker: Aalborg University (Denmark)
Released: 2019
Price: Not for sale
Status: Not for sale (research)
Best for: Professionals & labs
Programming: Open source (hardware + software)

Price and shipping

Master's thesis robot. All code is MIT-licensed on GitHub.

Shipping: N/A. Not for sale. The design files and code on GitHub are public, but no kit exists.

Specifications

Category	Spherical & rolling · Research & concepts
Released	2019
Country	Denmark
Availability	Not for sale (research)
Programmability	Open source (hardware + software)
Built-in autonomy	Navigates · Goes to places on its own, such as mapped rooms, waypoints or its charging dock
LLM support	Possible via SDK · No built-in LLM. Its SDK lets software on a computer control it.
MuJoCo simulation	None found · Nothing found.
NVIDIA Isaac simulation	None found · Nothing found.
Gazebo simulation	Sim-to-real kit · Maker's Kugle-Gazebo uses the same ROS topics and messages as the real robot's driver. Kugle-Gazebo ↗
Best for	Professionals & labs
Vision sensors	Lidar
Actuators	Unknown
Audio	No audio
Compute type	Unknown, Arduino / ESP32 / microcontroller
Remote control	Via SSH / web / bridge
Size	16 kg prototype (Kugle V1), roughly between waist and chest height with a tablet head (est.)
Locomotion	Three omni-wheels driving the ball
Compute	STM32H7 real-time board (FreeRTOS) plus an onboard ROS computer
Camera	Tablet display, no camera published
Sensors	IMU, encoders, two SICK TiM571 2D lidars
Runtime	Not published
Speed	Tested to 1 m/s and 1 rad/s
SDK	Open-source MATLAB/Simulink models, embedded firmware, ROS driver and Gazebo simulation (MIT)