Alexandre Mendes

About

Alexandre Mendes is a researcher whose work sits at the intersection of humanoid robotics, computer vision, and intelligent systems. His research has made meaningful contributions to the development of autonomous robotic platforms, particularly in the context of robot soccer — a demanding testbed for real-world AI and perception challenges. Among his most recognized contributions is his work on ball detection for humanoid robots, which has garnered 23 citations, alongside his development of NUClear, a loosely coupled, hybrid message-passing software architecture designed to reduce latency and improve modularity in humanoid robot systems. This architectural work reflects a broader commitment to building robust, scalable infrastructure for embodied AI. His investigations into colour modelling and unsupervised salient colour recognition demonstrate a sustained focus on making computer vision practical under real-time constraints, while his comparative analysis of deep learning computing platforms highlights an awareness of the hardware-software co-design challenges facing modern robotics. Mendes has also contributed to reinforcement learning for robot motor control and visual debugging tools, rounding out a research portfolio that spans both foundational and applied dimensions. With multiple publications in the respected *Intelligent Robotics and Applications* series, his work continues to serve as a valuable reference for researchers building intelligent, autonomous robotic systems.

Research Focus

Key Achievements

Top Papers

1. 1
   A Novel Approach to Ball Detection for Humanoid Robot Soccer

   23 citations · 2012
2. 2
   Intelligent Robotics and Applications

   21 citations · 2018
3. 3
   Intelligent Robotics and Applications

   20 citations · 2018
4. 4
   NUClear: A Loosely Coupled Software Architecture for Humanoid Robot Systems

   19 citations · 2016
5. 5
   Evaluation of Colour Models for Computer Vision Using Cluster Validation Techniques

   15 citations · 2013
6. 6
   Comparing Computing Platforms for Deep Learning on a Humanoid Robot

   14 citations · 2018
7. 7
   Unsupervised Recognition of Salient Colour for Real-Time Image Processing

   11 citations · 2014
8. 8
   Addressing the non-functional requirements of computer vision systems: a case study

   8 citations · 2015
9. 9
   Motivated Reinforcement Learning for Improved Head Actuation of Humanoid Robots

   7 citations · 2014
10. 10
    NUbugger: A Visual Real-Time Robot Debugging System

    7 citations · 2014

Key Collaborators

Contact & Links