Ezequiel Di Mario
École Normale Supérieure - PSL, École Polytechnique Fédérale de Lausanne
Papers
14
Total Citations
153
H-Index
8
About
Ezequiel Di Mario is a robotics and computational intelligence researcher whose work sits at the intersection of swarm robotics, evolutionary computation, and online learning. His primary contributions center on applying and adapting Particle Swarm Optimization (PSO) to the challenging problem of autonomous multi-robot controller design — particularly in noisy, real-world environments where traditional optimization methods struggle. Di Mario's most significant achievements involve developing distributed and noise-resistant variants of PSO that enable robots to learn cooperative behaviors, obstacle avoidance, and flocking dynamics efficiently and in limited time. His comparative analyses — benchmarking PSO against reinforcement learning and centralized versus distributed approaches — have provided the robotics community with practical guidelines for selecting optimization strategies in uncertain environments. His most cited work (29 citations) directly addresses the expensive, noisy nature of on-line robotic learning, a persistent challenge in the field. Beyond algorithmic contributions, Di Mario developed SwarmViz, an open-source visualization tool that makes PSO dynamics more interpretable and accessible to researchers. He also contributed a trajectory-based calibration method for stochastic motion models, demonstrating range across both theoretical and applied robotics. With a focused but impactful body of work accumulating over 130 citations, Di Mario represents an important voice in the growing field of adaptive swarm robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4SwarmViz: An open-source visualization tool for Particle Swarm Optimization14 citations · 2015
- 5
- 6A trajectory-based calibration method for stochastic motion models11 citations · 2011
- 7
- 8
- 9
- 10