TALOS

The walking humanoid robot that integrates the latest cutting-edge robotics technology

Torque control

The walking biped humanoid robot TALOS is fully torque-controllable thanks to the torque sensor feedback in all joints, which enables powerful sensing and multi-contact motions. The closed loop torque control can be used for whole-body control inverse dynamics and safe interaction with the environment.

TALOS-robot-communication-network-ethercat

EtherCAT Network

TALOS’ EtherCAT communication network allows to run control loops at 2 kHz, and up to 5 kHz, which enables TALOS to have highly reactive and dynamic motions.

EtherCAT Network

TALOS’ EtherCAT communication network allows to run control loops at 1 kHz, and up to 5 kHz, which enables TALOS to have highly reactive and dynamic motions.

6kg Payload per arm

TALOS has a payload of 6 Kg with each arm fully extended, because it was designed having in mind the use of heavy industrial tools. This enables our walking biped humanoid robot to effectively perform tasks in various environments such as the Factory of the Future.

Areas of use

USED FOR RESEARCH IN:

Artificial Intelligence/Machine Learning
Human-Robot Interaction
Manipulation
Perception
Navigation

Torque Control
Highly reactive and dynamic motion
Multi-contact motion planning
Dynamic Walking
Trajectory Optimization

FIELDS OF APPLICATION:

Factory of the Future

IoT

Rescue

Space exploration

TALOS IS USED IN…

Developing Memory of Motion in robots with arms and legs in project Memmo

Learn more >

TALOS, officially presented at LAAS-CNRS (Toulouse, France)

Learn more >

MEMMO Project: Giving memories to robots for a better motion generation

Learn more >

TALOS humanoid joins the RoboHub family at University of Waterloo!

Learn more >

TALOS at JSI: Humanoids, imitation learning and torque control

Learn more >

TALOS joins the robotics family at the University of Edinburgh

Learn more >

MEMMO: Giving memories to robots for a better motion generation

Learn more >

TALOS in action

TALOS is a high performance humanoid biped robot standing at 1.75m. PAL Robotics’ walking biped robot can lift up to 6 Kg with one stretched arm, is fully electrical and 100% ROS based.

TALOS in action
TALOS | The High-performance biped robot for research

TALOS in action

TALOS | The High-performance biped robot for research

Technical specifications

Contact PAL Robotics and get in touch with a team of experts
who can assess you and provide you with more information of TALOS.

Frontal view of the research humanoid biped robot TALOS

Side view of the humanoid research robot TALOS

SPECS

Height	175 cm
Weight	95 kg
Arm/Gripper payload	6 kg (arm stretched)
Battery autonomy	1.5h walking / 3h stand-by
Torque sensors	FULL TORQUE SENSOR FEEDBACK IN ALL JOINTS*
Head and Gripper	Fully customizable

*Except head, wrists and grippers

OPEN SOURCE

Simulation model available at: wiki.ros.org/Robots/TALOS

PREMIUM SOFTWARE

Speed your research with extra software: from Whole-Body Control
to navigation, HRI skills or teleoperation.

Download datasheet

Citations

Nahuel A. Villa, Pierre Fernbach, Maximilien Naveau, Guilhem Saurel, Ewen Dantec, Nicolas Mansard, Olivier Stasse (2022): Torque Controlled Locomotion of a Biped Robot with Link Flexibility

Tadej Petrič, Leon Žlajpah (2022): Phase State System for Generating Interactive Behaviors for Humanoid Robots

Noëlle Ramuzet (2022): Robotics Force/Torque Control for Manufacturing Operations

Alexis Nicolin (2022): Planning of visual servoing tasks for robotics

Kevin Giraud-Esclasse (2022): Towards a reactive motion generation on exteroceptive feedback for generalized locomotion of humanoid robots

Sébastien Kleff, Ewen Dantec, Guilhem Saurel, Nicolas Mansard, Ludovic Righetti (2022): Introducing Force Feedback in Model Predictive Control

Noëlle Ramuzat (2022): Contrôle en force/couple pour des opérations industrielles

Noëlie Ramuzat, Olivier Stasse, S. Boria (2022): Benchmarking Whole‐Body Controllers on the TALOS Humanoid Robot

Evelyn D’Elia, Jean-Baptiste Mouret, Jens Kober, Serena Ivaldi (2022): Automatic Tuning and Selection of Whole-Body Controllers

Ewen Dantec, Michel Taïx, Nicolas Mansard (2022): First Order Approximation of Model Predictive Control Solutions for High Frequency Feedback

Stylianos Piperakis, Michael Maravgakis, Dimitrios Kanoulas, Panos Trahanias (2022): Robust Contact State Estimation in Humanoid Walking Gaits

Zvezdan Lončarević, Tadej Petrič, Andrej Gams (2022): Fitting Constrained Trajectory with High Variability into Redundant Robot Workspace

Ewen Louis Dantec, Maximilien Naveau, Nicolas Mansard, Pierre Fernbach, Nahuel Villa, Guilhem Saurel, Olivier Stasse, Michel Taïx (2022): Whole-Body Model Predictive Control for Biped Locomotion on a Torque-Controlled Humanoid Robot

Evelyn D’Elia, Jean-Baptiste Mouret, Jens Kober, Serena Ivaldi (2022): Learning task controllers on a humanoid robot using multi-objective optimization

Eloïse Dalin, Ivan Bergonzani, Timothée Anne, Serena Ivaldi, Jean-Baptiste Mouret (2022): First Do Not Fall: Learning to Exploit a Wall With a Damaged Humanoid Robot

Ramón Calvo González (2021): Planificación y seguimiento de trayectorias aplicado al movimiento de robots bípedos

Ubieta Paéz, Ignacio de Loyola (2021): Diseño, simulación y control de un robot humanoide espacial

Ahmad Gazar, Majid Khadiv, Andrea Del Prete, Ludovic Righetti (2021): Stochastic and Robust MPC for Bipedal Locomotion: A Comparative Study on Robustness and Performance

Yanlei Ye, Peng Li, Zihao Li, Fugui Xie, Xin-Jun Liu, Jianhui Lie (2021): Real-Time Design Based on PREEMPT_RT and Timing Analysis of Collaborative Robot Control System

Guillaume Gobin (2021): Localisation d’un robot humanoïde TALOS par information géométrique et visuelle combinée à l’apprentissage

Evelyn D’Elia (2021): Automatic Controller Selection on a Humanoid Robot

Ewen Dantec, Rohan Budhiraja, Adrià Roig, Teguh Lembono, Guilhem Saurel, Olivier Stasse, Pierre Fernbach, Steve Tonneau, Sethu Vijayakumar, Sylvain Calinon, Michel Taïx, Nicolas Mansard (2021): Whole Body Model Predictive Control with a Memory of Motion: Experiments on a Torque-Controlled Talos

Zvezdan Lončarević, Rok Pahič, Aleš Ude, Andrej Gams (2021): Generalization-Based Acquisition of Training Data for Motor Primitive Learning by Neural Networks

Tianye Tan, Chengyu Cui, Sungkwon Park (2021): The In-Robot Network Structure of Humanoid Robot for Burst Traffic Data Situations

Isabelle Maroger, Noëlie Ramuzat, Olivier Stasse, Bruno Watier (2021): Human Trajectory Prediction Model and Its Coupling With a Walking Pattern Generator of a Humanoid Robot

Kristina Savevska, Mihael Simonič, Aleš Ude (2021): Modular Real-Time System for Upper-Body Motion Imitation on Humanoid Robot Talos

Teguh Santoso Lembono, Emmanuel Pignat, Julius Jankowski, Sylvain Calinon (2021): Learning Constrained Distributions of Robot Configurations With Generative Adversarial Network

Isabelle Maroger, Olivier Stasse, Bruno Watier (2021): Walking Human Trajectory Models and Their Application to Humanoid Robot Locomotion

Aghil Yousefi-Koma, Behnam Maleki, Hessam Maleki, Amin Amani, Mohammad Ali Bazrafshani, Hossein Keshavarz, Ala Iranmanesh, Alireza Yazdanpanah, Hamidreza Alai, Sahel Salehi, Mahyar Ashkvari, Milad Mousavi, Milad Shafiee Ashtiani (2021): SURENAIV: Towards A Cost-effective Full-size Humanoid Robot for Real-world Scenarios

N. Ramuzat, G. Buondonno, S. Boria, O. Stasse (2021): Comparison of Position and Torque Whole-Body Control Schemes on the Humanoid Robot TALOS

T. Lasguignes, I. Maroger, M. Fallon, M. Ramezani, L. Marchionni, O. Stasse, N. Mansard, B. Watier (2021): ICP Localization and Walking Experiments on a TALOS Humanoid Robot

Eloïse Dalin, Ivan Bergonzani, Timothée Anne, Serena Ivaldi, Jean-Baptiste Mouret (2021): Whole-body teleoperation of the Talos humanoid robot: preliminary results

Luka Mišković, Rebeka Kropivšek, Andrej Gams, Tadej Petrič (2021): Optimizing end-effector force during the sit-stand task on the Talos humanoid bipedal robot

Daeun Song, Pierre Fernbach, Thomas Flayols, Andrea Del Prete, Nicolas Mansard, Steve Tonneau, Young J. Kim (2021): Solving Footstep Planning as a Feasibility Problem Using L1-Norm Minimization

Noélie Ramuzat, Florent Forget, V. Bonnet, S Boria, Olivier Stasse, Maxime Gautier (2020): Actuator Model, Identification and Differential Dynamic. Programming for a TALOS Humanoid Robot

Alexis Nicolin, Joseph Mirabel, Sébastien Boria, Olivier Stasse, Florent Lamiraux (2020): Agimus: a new framework for mapping manipulation motion plans to sequences of hierarchical task-based controllers. DOI: 10.1109/SII46433.2020.9026288

Kevin Giraud-Esclasse, Pierre Fernbach, Gabriele Buondonno, Carlos Mastalli and Olivier Stasse (2020): Motion Planning for Multi-Contact Visual Servoing on Humanoid Robots. DOI: 10.1109/SII46433.2020.9026291

Steve Tonneau, Daeun Song, Pierre Fernbach, Nicolas Mansard, Michel Taix, Andrea Del Prete (2020): SL1M: Sparse L1-norm Minimization for contact planning on uneven terrain

Rohan Budhiraja, Amit Parag, Ewen Dantec, Justin Carpentier, Carlos Mastalli, Nicolas Mansard (2020): Crocoddyl: Fast computation, Efficient solvers, Receding horizon and Learning

Teguh Santoso Lembono, Emmanuel Pignat, Julius Jankowski, Sylvain Calinon (2020): Generative Adversarial Network to Learn Valid Distributions of Robot Configurations for Inverse Kinematics and Constrained Motion Planning

Louise Scherrer (2020): Reactive Walking for the Humanoid Robot Pyrène

Josep Martí-Saumell, Joan Solà, Carlos Mastalli, Angel-Santamaría-Navarro (2020): Squash-Box Feasibility Driven Differential Dynamic Programming

Zvezdan Lončarević, Rok Pahič, Aleš Ude, Andrej Gams (2020): Reduction of Trajectory Encoding Data using a Deep Autoencoder Network: Robotic Throwing

Santoso Lembono, Teguh Mastalli, Carlos Fernbach, Pierre Mansard, Nicolas Calinon, Sylvain (2020): Learning How to Walk: Warm-starting Optimal Control Solver with Memory of Motion

Carlos Mastalli, Rohan Budhiraja, Wolfgang Merkt, Guilhem Saurel, Bilal Hammoud, Maximilien Naveau, Justin Carpentier, Ludovic Righetti, Sethu Vijayakumar, Nicolas Mansard (2020): Crocoddyl: An Efficient and Versatile Framework for Multi-Contact Optimal Control

Barry Ridge, Rok Pahič, Aleš Ude, Jun Morimoto (2019): Learning to Write Anywhere with Spatial Transformer. Image-to-Motion Encoder-Decoder Networks. DOI: 10.1109/ICRA.2019.8794253

Tamim Asfour, Mirko Wächter, Lukas Kaul, Samuel Rader, Pascal Weiner, Simon Ottenhaus, Raphael Grimm, You Zhou, Markus Grotz, and Fabian Paus (2019): ARMAR-6: A High-Performance Humanoid for Human-Robot Collaboration in Real-World Scenarios. DOI: 10.1109/MRA.2019.2941246

Emmanuel Pignat, Teguh Lembono and Sylvain Calinon (2019): Variational Inference with Mixture Model Approximation: Robotic Applications

Olivier Stasse, Thomas Flayols, Rohan Budhiraja, Kevin Giraud-Esclasse, Justin Carpentier, Joseph Mirabel, Andrea del Prete, Philippe Souères, Nicolas Mansard, Florent Lamiraux (2018): TALOS: A new humanoid research platform targeted for industrial applications. DOI: 10.1109/HUMANOIDS.2017.8246947

Olivier Stasse and Thomas Flayols (2018): An overview of humanoid robots technologies

Gopi Krishna Erabati (2018): 3D Object Recognition and Relative. Localization using a 3D sensor Embedded on a Mobile Robot. DOI: 10.31237/osf.io/p4thn

Justin Carpentier, Nicolas Mansard (2018): Multi-contact Locomotion of Legged Robots

O. Stasse, J. Mirabel, A. Nicolin and F. Lamiraux1 (2017): Research results on the robot TALOS

Justin Carpentier, Rohan Budhiraja, Nicolas Mansard (2017): Learning Feasibility Constraints for Multi-contact. Locomotion of Legged Robots

Justin Carpentier, Andrea del Prete, Steve Tonneau, Thomas Flayols, Florent Forget, Alexis Mifsud, Kevin Giraud–Esclasse, Dinesh Atchuthan, Pierre Fernbach, Rohan Budhiraja (2017): Multi-contact Locomotion of Legged Robots in Complex Environments – The Loco3D project

Sachin Chitta, Eitan Marder-Eppstein, Wim Meeussen, Vijay Pradeep, Adolfo Rodríguez Tsouroukdissian, Jonathan Bohren, David Coleman, Bence Magyar, Gennaro Raiola, Mathias Lüdtke (2017): ros_control: A generic and simple control framework for ROS. DOI : 10.21105/joss.00456

TALOS Research Projects in Humanoid and Biped Robotics

Frequently Asked Questions

Which is the frequency of the control loop?

TALOS’ Torque control has a low latency of 2kHz.

Is there a full state estimation of the robot available to the controllers?

We provide a state estimation from IMU+kinematics out-of-the-box, a system for fusing this odometry with additional external measurements should be tackled with an additional estimator out of the real-time ros_control loop. This would be safer for the whole body torque control that rely on the CoM estimation.

Which sensors are available through ros_control?

Via ros_control, you have access to the encoders, IMU, torque sensors at a joint level, Force/Torque sensors on the wrists and anckles and temperature sensors of the robots.

Is the dynamical model of the robot provided?

Yes, the URDF includes CoM position, mass and inertia tensors. This model has been extracted from the CAD and it was validated by the full body inverse dynamics and torque control on the walking biped robot.

More FAQs

Contact PAL Robotics team if you have any further questions!

Subscribe to our Newsletter

Register for PAL Robotics’s newsletter and stay up to date with product news, relevant events and exhibitions, and other relevant information about PAL Robotics.