Pronto is an efficient, versatile and modular EKF state estimator for both proprioceptive (inertial, kinematics) and exteroceptive (LIDAR, camera) sensor fusion. It has been used with a variety of quadruped and humanoid robots.

It supports sensors such as as IMUs (Microstrain, KVH, XSense), LIDAR (Hokuyo, Velodyne, Hesai), cameras and joint kinematics.

It was originally developed for the Boston Dynamics Atlas robot but has been heavily developed to work with quadrupeds in particular the ANYbotics ANYmal.

Updates

November 2025: Pronto has been updated to ROS2 with the support of Francesco Iotti and colleagues in Università di Pisa. The main branch of the repo targets ROS 2 Humble. It has been tested with logs from Anymal version D.

Check out the repo on GitHub:

https://github.com/ori-drs/pronto

Pronto

Publications

If you use Pronto in your work, please cite the following publication:

M. Camurri, M. Ramezani, S. Nobili, M. Fallon Pronto: A Multi-Sensor State Estimator for Legged Robots in Real-World Scenarios in Frontiers on Robotics and AI, 2020 (PDF) DOI: 10.3389/frobt.2020.00068

@article{camurri2020frontiers,
  author = {Camurri, Marco and Ramezani, Milad and Nobili, Simona and Fallon, Maurice},
  title = ,
  journal = {Frontiers in Robotics and AI},
  volume = {7},
  number = {68},
  pages = {1--18},
  year = {2020},
  url = {https://www.frontiersin.org/article/10.3389/frobt.2020.00068},
  doi = {10.3389/frobt.2020.00068},
  issn = {2296-9144}
}