Diego Cabrera

About

Diego Cabrera is a prominent researcher specializing in intelligent fault diagnosis, generative adversarial networks (GANs), and industrial robotics condition monitoring. His work addresses one of the most persistent challenges in data-driven machine learning: the scarcity and imbalance of fault data in real-world industrial environments, particularly for robotic manipulators and transmission systems. Cabrera's most impactful contribution lies in pioneering GAN-based frameworks for fault diagnosis under data-limited conditions. His 2021 paper introducing a One-Class Generative Adversarial Detection Framework (55 citations) demonstrated how unbalanced datasets could be effectively handled through generative modeling. Building on this, he developed VGAN, a principled generalization of MSE GAN and WGAN-GP (23 citations), and explored Sliced Wasserstein cycle-consistency architectures for data augmentation (22 citations). His earlier work on GANs as oversampling tools for robotic manipulator diagnostics (22 citations) helped establish the foundational methodology that subsequent studies have built upon. More recently, Cabrera has pushed toward few-shot and one-shot learning paradigms, enabling fault diagnosis systems to operate effectively with minimal labeled examples — a critical advancement for practical industrial deployment. With roots in embedded robotics systems, his research trajectory reflects a deep, evolving commitment to making intelligent maintenance solutions robust, generalizable, and industrially viable.

Research Focus

Key Achievements

Top Papers

1. 1
   A One-Class Generative Adversarial Detection Framework for Multifunctional Fault Diagnoses

   55 citations · 2021
2. 2
   VGAN: Generalizing MSE GAN and WGAN-GP for Robot Fault Diagnosis

   23 citations · 2022
3. 3
   Sliced Wasserstein cycle consistency generative adversarial networks for fault data augmentation of an industrial robot

   22 citations · 2023
4. 4
   Exploiting Generative Adversarial Networks as an Oversampling Method for Fault Diagnosis of an Industrial Robotic Manipulator

   22 citations · 2020
5. 5
   Generative adversarial one-shot diagnosis of transmission faults for industrial robots

   19 citations · 2023
6. 6
   Incrementally Generative Adversarial Diagnostics Using Few-Shot Enabled One-Class Learning

   18 citations · 2024
7. 7
   Balance of a hexapod in real time using a FPGA

   2 citations · 2015

Key Collaborators

Contact & Links