Mirko Ferrati

I developed a planning scheme which aims to solve the manipulation tasks for robotic manipulators using environment-fixed support surfaces and mobile robots as well. The tasks of the robot are broken down into elementary manipulation blocks. Finally, a time-varying graph is built in order to handle the multiple objects in dynamic chains of robots.
More can be seen in the YouTube video here: Youtu.be/vq10SvHYzsk

I designed most of the software architecture used to control and teleoperate the Walk-Man robot.

The architecture included multiple operators and onboarding computers, along with tens of distributed process interconnected by two different communication middlewares: one for high-frequency control on the robot and one for the low-frequency teleoperation and visualization by the human operators.

Such processes were developed by a team of more than ten robotics researchers and engineers, in a common framework with multiple shared libraries and utilities. More information can be found at the YouTube video here: Youtu.be/94ngO1nKXgs

This is one of the latest planning algorithms for controlling multiple robots that I developed, the planner itself (this package) needs to be used with a simulator or with real robots. However, it's still a good sample of a high-level application that uses a middleware (ROS) for communication between different nodes and uses a complex C++ planner optimized for speed (since the robots are moving while planning).

The website describes how to use the whole system in the simulation, but requires some prior ROS knowledge and external dependencies.

This is a scientific paper that I co-wrote concerning a simulator for mobile robots.
Authors: M. Ferrati, A. Settimi, and L. Pallottino
Written for the International Workshop on Modelling and Simulation for Autonomous Systems.

A scientific publication about a system that remotely operates a robot in disasters using a custom Qt GUI.