Max Robotics — Global Service Robot Directory + US Certification

About

Davide Azzalini is a robotics and artificial intelligence researcher whose work centers on autonomous systems, anomaly detection, and multi-agent coordination. His research addresses one of the most pressing challenges in long-term robot autonomy: enabling robots to reliably identify faults and anomalous behaviors without requiring extensive labeled training data. Azzalini's most impactful contribution is a minimally supervised anomaly detection framework leveraging Variational Autoencoders, designed to model complex, long multivariate time-series data from autonomous robots. Published in 2021, this work has garnered 35 citations, establishing it as a notable reference in the field. Complementing this, his earlier investigations into Hidden Markov Models (HMMs) for anomaly detection provided foundational probabilistic approaches to characterizing and comparing robot behaviors, themes that also form the backbone of his doctoral research. More recently, Azzalini has expanded his scope to multi-agent systems, tackling the Multi-Agent Pickup and Delivery (MAPD) problem in dynamic environments, with a particular focus on deadlock prevention — a critical challenge for deploying robot fleets in real-world settings. Across his body of work, Azzalini consistently bridges theoretical machine learning methods with practical robotics applications, making meaningful contributions toward safer, more resilient autonomous systems.