Jonathan Piat

About

Jonathan Piat is a researcher at the forefront of embedded computer vision and hardware-software co-design, with a focus on enabling real-time robotic perception on resource-constrained platforms. His work centers on accelerating Simultaneous Localization and Mapping (SLAM) algorithms—a critical capability for autonomous navigation—using Field-Programmable Gate Arrays (FPGAs). Piat’s major contributions include the development of a complete HW/SW co-design framework for visual SLAM, demonstrating how to offload computationally intensive tasks like feature extraction and matching onto reconfigurable hardware. His 2018 paper on this topic (8 citations) provides a blueprint for achieving real-time performance in monocular SLAM systems. Earlier, he designed a real-time FPGA-based vision system for obstacle detection and localization (7 citations), showcasing practical applications in robotics. Piat also pioneered the hardware acceleration of a BRIEF correlator module (5 citations), a key component for efficient feature matching in SLAM. His work bridges the gap between algorithmic complexity and embedded deployment, making robot localization more accessible for real-world applications.

Research Focus

Key Achievements

Top Papers

1. 1
   HW/SW co-design of a visual SLAM application

   8 citations · 2018
2. 2
   Real Time Vision System for Obstacle Detection and Localization on FPGA

   7 citations · 2015
3. 3
   FPGA based hardware acceleration of a BRIEF correlator module for a monocular SLAM application

   5 citations · 2016

Key Collaborators

Contact & Links