HomeRobots
BRUCE

BRUCE

BRUCE by Westwood Robotics is a 70cm, 4.8kg kid-size humanoid robot with 16 DOF, liquid-cooled actuators, and open-source design for research and education.
In Development
USA
Humanoid
2023
Specifications
Dev Doc
Website
Made by
Westwood Robotics Corporation
Software Type
Open Source
Software Package
Open-source control software. Support for mainstream deep learning frameworks. Model Predictive Control (MPC) algorithms for dynamic locomotion. Wireless control interfaces
Actuators
BRUCE uses Koala BEAR liquid-cooled actuators, especially in the knees, providing high burst torque (8 Nm) and dynamic performance with efficient thermal management.
Compiute
Equipped with a main computer delivering up to 6 TOPS of computing power, 8GB RAM, and 32GB storage, capable of running complex AI and control algorithms in real time.
Sensors
4 contact sensors for ground and object interaction 6-DOF inertial measurement unit (IMU) sampling at 2 kHz for balance and motion sensing
Max Op. time
mins

Robot Brief

BRUCE (Bipedal Robot Unit with Compliance Enhanced) is a kid-size humanoid robot developed collaboratively by Westwood Robotics and RoMeLa at UCLA, designed as an open-platform for robotics research and education. Constructed using a lightweight carbon fiber composite structure, BRUCE weighs only 4.8 kg and stands 70 cm tall, making it highly portable and suitable for dynamic research environments. Equipped with 16 degrees of freedom, including 5 DOF per leg and 3 DOF per arm, BRUCE features advanced liquid-cooled knee actuators that enable explosive power and dynamic motions such as walking, running, and jumping. Its intelligent navigation and sensor suite—including inertial measurement units and contact sensors—allow it to adapt in real time to environmental changes. The robot supports open-source software and hardware design, facilitating easy repair, modification, and experimentation. With a 3000mAh battery providing about 20 minutes of continuous dynamic operation, BRUCE offers researchers a robust, modular, and highly capable platform for advancing humanoid robotics.

Use Cases

BRUCE performs dynamic bipedal locomotion including walking, running, and jumping. It autonomously navigates environments, adapts to changes using sensor feedback, and supports complex motion planning. Designed primarily for research and education, it enables experimentation with advanced control algorithms, proprioceptive actuation, and human-robot interaction.

Industries

  • Robotics Research: Provides an open platform for developing and testing humanoid locomotion and control algorithms.
  • Education: Used in universities and labs for hands-on learning in robotics and AI.
  • Industrial R&D: Supports prototyping and development of dynamic humanoid systems.
  • Entertainment (suggested): Potential for dynamic demonstrations and interactive robotics.

‍

Specifications

Length
mm
Width
mm
Height (ResT)
mm
Height (Stand)
-
700
mm
Height (Min)
mm
Height (Max)
mm
Weight (With Batt.)
-
kg
Weight (NO Batt.)
-
4.8
kg
Max Step Height
-
mm
Max Slope
+/-
-
°
Op. Temp (min)
-
°C
Op. Temp (Max)
-
°C
Ingress Rating
-
No items found.

Intro

BRUCE is a compact humanoid robot featuring a carbon fiber composite body for lightweight durability. It has 16 degrees of freedom distributed across its legs and arms, with high-performance liquid-cooled knee actuators enabling dynamic motions like jumping. The robot’s sensor suite includes 4 contact sensors and a 6-DOF inertial measurement unit sampling at 2 kHz, providing precise feedback for balance and motion control. Its main computer delivers up to 6 TOPS of computing power with 8GB RAM and 32GB storage, supporting mainstream deep learning frameworks. Controlled wirelessly via LAN or Bluetooth, BRUCE offers researchers an open-source hardware and software platform with modular components that are easy to repair and customize.

Connectivity

  • Wireless LAN (Wi-Fi) with SSH control
  • Bluetooth connectivity
  • USB ports for data and power interfaces

‍

Capabilities

  • Dynamic walking, running, and jumping enabled by liquid-cooled actuators
  • 16 degrees of freedom for versatile movement
  • Real-time sensor feedback with high-frequency IMU and contact sensors
  • Wireless control via LAN and Bluetooth
  • Open-source software supporting mainstream deep learning frameworks
  • Modular, repairable design for rapid prototyping and experimentation

‍