Juan Andrade Cetto

I'm a Research Scientist at the Institut de Robòtica i Informàtica Industrial, CSIC-UPC. My research focuses on robot perception and learning, with particular emphasis on visual SLAM, active exploration, event-based vision, and aerial manipulation. I lead projects developing novel solutions for robot mapping, localization, and navigation using advanced sensing technologies like event-based cameras and laser rangefinders. My work bridges fundamental theory with practical applications, from developing information-theoretic frameworks for robot exploration to building real-time systems for autonomous aerial and ground robots. I am particularly interested in pushing the boundaries of robot perception through novel sensor fusion approaches and machine learning techniques. My group has made significant contributions in areas like active SLAM with information-theoretic exploration strategies, event-based visual processing for high-speed robot motion, and visual servoing for aerial manipulation. We maintain strong collaborations with leading robotics groups worldwide and regularly validate our research on real robot platforms.

Publications

Egomotion from event-based SNN optical flow

Yi Tian, J. Andrade-Cetto

International Conference on Systems 2023

Juan Andrade Cetto

Publications

Egomotion from event-based SNN optical flow

Borinot: an open thrust-torque-controlled robot for research on agile aerial-contact motion

Event-IMU fusion strategies for faster-than-IMU estimation throughput

Control‐oriented estimation of the exchange current density in PEM fuel cells via stochastic filtering

Spiking Neural Network for Event camera ego-motion estimation

Event-Based Line SLAM in Real-Time

Dual-Branch CNNs for Vehicle Detection and Tracking on LiDAR Data

Wolf: A Modular Estimation Framework for Robotics Based on Factor Graphs

Full-Body Torque-Level Non-linear Model Predictive Control for Aerial Manipulation

High-speed event-based camera tracking

Multi-task closed-loop inverse kinematics stability through semidefinite programming

Active SLAM for Autonomous Underwater Exploration

Timed-Elastic Smooth Curve Optimization for Mobile-Base Motion Planning

Joint on-manifold self-calibration of odometry model and sensor extrinsics using pre-integration

LOGIMATIC Autonomous Navigation

Pose-graph SLAM sparsification using factor descent

Visual Guidance of Unmanned Aerial Manipulators

The AEROARMS Project: Aerial Robots with Advanced Manipulation Capabilities for Inspection and Maintenance

Boosted Random Ferns for Object Detection

Graph SLAM Sparsification With Populated Topologies Using Factor Descent Optimization

Autonomous navigation of micro aerial vehicles using high-rate and low-cost sensors

Low Resolution Lidar-Based Multi-Object Tracking for Driving Applications

Deconvolutional networks for point-cloud vehicle detection and tracking in driving scenarios

Factor descent optimization for sparsification in graph SLAM

Mapping, Planning and Exploration with Pose SLAM

Nonlinear model predictive control for aerial manipulation

Trajectory Generation for Unmanned Aerial Manipulators Through Quadratic Programming

Uncalibrated Visual Servo for Unmanned Aerial Manipulation

Word Ordering and Document Adjacency for Large Loop Closure Detection in 2-D Laser Maps

Observability analysis and optimal sensor placement in stereo radar odometry

MSClique: Multiple Structure Discovery through the Maximum Weighted Clique Problem

High-frequency MAV state estimation using low-cost inertial and optical flow measurement units

Uncertainty analysis of the DLT-Lines calibration algorithm for cameras with radial distortion

Potential information fields for mobile robot exploration

Active pose SLAM with RRT*

Terrain Classification in Complex Three‐dimensional Outdoor Environments

LETHA: Learning from High Quality Inputs for 3D Pose Estimation in Low Quality Images

Task priority control for aerial manipulation

Dense entropy decrease estimation for mobile robot exploration

Exploration on continuous Gaussian process frontier maps

Localization in highly dynamic environments using dual-timescale NDT-MCL

Calibration of an Outdoor Distributed Camera Network with a 3D Point Cloud

HRA∗: Hybrid randomized path planning for complex 3D environments

Exhaustive Linearization for Robust Camera Pose and Focal Length Estimation

Estimation of camera calibration uncertainty using LIDAR data

Mobile robot exploration with potential information fields

Planning Reliable Paths With Pose SLAM

Uncalibrated image-based visual servoing

Active Pose SLAM

Bootstrapping Boosted Random Ferns for discriminative and efficient object classification

Dense outdoor 3D mapping and navigation with Pose SLAM

Detection Performance Evaluation of Boosted Random Ferns

Path planning in belief space with pose SLAM

Amortized constant time state estimation in Pose SLAM and hierarchical SLAM using a mixed Kalman-information filter

Action Selection for Single-Camera SLAM

Shared Random Ferns for Efficient Detection of Multiple Categories

Efficient rotation invariant object detection using boosted Random Ferns

Object modeling using a ToF camera under an uncertainty reduction approach

Decentralized Sensor Fusion for Ubiquitous Networking Robotics in Urban Areas

Information-Based Compact Pose SLAM

Reduced state representation in delayed-state SLAM

3D mapping for urban service robots

Calibrating an outdoor distributed camera network using Laser Range Finder data

Humanoid robotics and human-centered initiatives at IRI

Local Boosted Features for Pedestrian Detection

Combining color-based invariant gradient detector with HoG descriptors for robust image detection in scenes under cast shadows

Recovering epipolar direction from two affine views of a planar object

A Wire-Based Active Tracker

Vision-based loop closing for delayed state robot mapping

Robust Color Contour Object Detection Invariant to Shadows

Unidimensional Multiscale Local Features for Object Detection Under Rotation and Mild Occlusions

On the Observability of Bearing-only SLAM

Orientation Invariant Features for Multiclass Object Recognition

Computing the rate of spread of linear flame fronts by thermal image processing

Affine Epipolar Direction from Two Views of a Planar Contour

Computation of Rotation Local Invariant Features using the Integral Image for Real Time Object Detection

Environment Learning for Indoor Mobile Robots: A Stochastic State Estimation Approach to Simultaneous Localization and Map Building (Springer Tracts in Advanced Robotics)

Active control for single camera SLAM