My research focuses on robot perception, deep learning, sensor-based motion planning, and control systems. What I love most are computer vision, 3D SLAM (Simultaneous Localization And Mapping), and spatial AI for their exciting mix of geometry, state estimation, optimization, deep learning, and many other beautiful mathematical tools. 

RGB-D and Stereo SLAM: Volumetric reconstruction and 3D incremental segmentation 
PLVS is a real-time system that leverages sparse RGB-D and Stereo SLAM, volumetric mapping, and 3D unsupervised incremental segmentation. PLVS stands for Points, Lines, Volumetric mapping, and Segmentation. You can find further details on this page


Monocular SLAM with modern local features

 pySLAM is a Python framework that allows you to play with SLAM techniques, feature-matching, and many modern local features (based on new Deep Learning approaches). All the supported local features are accessible within a single Python environment. See this page for further information. 

Perception and motion planning  for UGVs
Point cloud segmentation and 3D mapping by using RGBD and LIDAR sensors 
• Sensor-based motion planning in challenging scenarios

DortmundPhoenix Traversability

Learning Algorithms for UGVs
Online WiFi mapping by using Gaussian processes
Robot dynamics/kinematics learning and control (algorithms for learning the model of UGVs dynamics/kinematics)


Multi-robot systems
3D Patrolling with a team of UGVs
3D Exploration with a team of UGVs
Coverage with a heterogeneous team of robots
• Exploration with mobile robots: the Multi-SRG Method


Computer vision and visual servoing for UAVs
During my experience in Selex ES MUAS (now Leonardo), I designed and developed advanced real-time applications and managed applied research projects for UAVs (CREX-BASIOBSPYBALL-B). In particular, one of these projects focused on near real-time 3D reconstruction from a live video stream. See more on my LinkedIn profile


Computer vision and visual servoing for mobile robots
Intercepting a moving object with a mobile robot
Appearance-based nonholonomic navigation from a database of images


Exploration of unknown environments
• With mobile robots: the SRT Method
• With a multi-robot system: the Multi-SRG Method
• With a general robotic system: the Sensor-based Exploration Tree

 SRG-sim    SET-3sensors-6+3

Robot navigation and mapping with limited sensing
In this work, we characterize the information space of a robot moving in a plane with limited sensing. The robot has a landmark detector, which provides the cyclic order of the landmarks around the robot, and it also has a touch sensor, that indicates when the robot is in contact with the environment boundary. The robot cannot measure any precise distances or angles and does not have an odometer or a compass. We propose to characterize the information space associated with such a robot through the swap cell decomposition. We show how to construct such decomposition through its dual, called the swap graph, using two kinds of feedback motion commands based on the landmarks sensed. See more in these papers
Learning Combinatorial Map Information from Permutations of Landmarks
Learning combinatorial information from alignments of landmarks
Using a Robot to Learn Geometric Information from Permutations of Landmarks



Warning: The material (PDF or multimedia) that can be downloaded from this page may be subject to copyright restrictions. Only personal use is allowed.