Humanoid Robot - Toala

Toala is an ideal robot for developing full-body humanoid robot applications.

Features

Size and strength equivalent to humans

With a height of 1300~1640 mm (variable), a reach of 740 mm, and a cart width of 720 mm, the size of the robot is close to that of humans. The payload of one arm is 6 kg at the worst-case holding posture, which is sufficient to develop many applications.

Sufficient joint configuration for a wide variety of tasks

The joint configuration of 7-axis dual arms, 3-axis waist (yaw, pitch, elevation), 2-axis neck (yaw, pitch), and 4-axis undercarriage (omni-directional mobile base) enables the robot to do tasks with a range of motion similar to human beings in living and working spaces of people.

Torque sensing of arm and waist

By using joint disturbance observers for the arms and waist, external force detection is possible without torque sensors, enabling contact stop and joint flexibility.

ROS support

Because it is ROS-compatible, the robot in Gazebo (a simulator for ROS) and the actual robot can be operated using the same program. This makes it possible to safely verify the robot's behavior. In addition, trajectory planning and self-interference detection using MoveIt! are implemented as standard.

Movies

Basic Motion of Humanoid Robot Toala

This is a demo of Toala's whole body movement. The second half shows joint impedance control of the arms and waist.

Remote Control of Humanoid Robot Toala

A remote control demo using Noitom's Perception Neuron PRO.

Specifications

Sensors (head)
  • Depth camera x 1
  • Stereo microphone x 1
  • Wide-angle stereo camera x 1 (optional)
  • Fisheye camera x 1 (optional)
Effectors (head)
  • Speaker x 1
Height 1300 ~ 1640 mm
Weight approx. 120 kg
Payload Worst posture: 6 kg (single arm)
External terminals (back of the robot)
  • Digital input x 7
  • Digital output x 7
  • Analog input: x 8
  • Analog output x 2
  • Power supply for external equipment: 5V-DC/2A, 12V-DC/2A, 24V-DC/2A
  • Ethernet x 1
  • USB 3.0 x 3
  • HDMI output x 1
  • Charger connection terminal
Tool flange (tip of the arm)
  • 24V DC power supply terminal x 1
  • General-purpose signal wire x 8
Onboard computer
  • PC for robot control
  • PC for image acquisition, etc.
Power source Built-in battery (continuous operation while charging is also possible)
Battery continuous operation time Up to approx. 3 hours

Dimensions

 

Software

The basic functions required to control a full-body humanoid robot are implemented. Toala can utilize joint impedance control based on torque detection, fall prevention by monitoring ZMP, a state machine for connecting multiple movements, and safety stop based on interference detection. Toala's software is ROS compatible. Therefore, in addition to the easy use of the above functions, state visualization with RViz, trajectory planning using a ROS standard software MoveIt!, and logging and saving of the robot's sensor information (camera images, joint angles, joint torques, etc.) are available.

Basic Package and Optional Extras

Basic Package

  • Humanoid robot - Toala
  • Battery charger
  • ROS package for Toala
  • User's manual
  • Six months email support

 

Upper Body Package

  • Upper body humanoid robot - Toala (without mobile base and elevator)
  • Control PC
  • Battery charger
  • ROS package for Toala
  • User's manual
  • Six months email support

 

Force-torque Sensors

This option enables advanced force control at the tip of the arm. A low-profile 6-axis force-torque sensor and an impedance control program are provided.

 

Hands

Three-finger Hand

  • Three finger configuration: thumb + 2 fingers
  • One axis for each finger (3 motors in total)
  • Outputting about 10N force at fingertips
  • Rubber material attached to fingertips and palm

 

Four-finger Hand

  • Impedance control at each joint
  • Four-finger configuration: thumb + 3 fingers
  • Three axes for each finger (10 actuated and 3 passive joints)
  • Outputting about 8N force at fingertips
  • Rubber material and FSR pressure sensor attached to fingertips and palm
  

 

Leader-follower Control System

Noitom's motion capture system, Perception Neuron® PRO, is used. This system can be used for motion teaching for machine learning as well as remote control applications. The system provides singularity avoidance, self-interference protection, and a torque-limiting function when the robot is physically constrained (e.g., when pushing a desk from above). It also has functions to enable/disable control of the left and right arms, grippers, neck, waist, and mobile base independently.

Mapping of Operator and Robot

Software