Max Robotics — Global Service Robot Directory + US Certification

About

Duvindu Piyasena is a researcher at the forefront of lifelong learning and edge computing for autonomous systems. His work focuses on enabling robots to continuously learn and adapt from their environment, a capability essential for real-world deployment. He made significant contributions to the IROS 2019 Lifelong Robotic Vision Challenge, where his team was among the top finalists out of over 150 teams, helping to establish the OpenLORIS benchmark for lifelong object recognition. Piyasena has pioneered hardware acceleration for lifelong deep learning, designing an FPGA accelerator that uses streaming Linear Discriminant Analysis to achieve real-time performance on edge devices—a critical advancement for drones and robotics. His work on a binary search tree-based feature matching accelerator further addresses the challenge of real-time localization with limited onboard computing. With over 20 citations across his most influential papers, Piyasena’s research bridges the gap between algorithmic lifelong learning and practical, energy-efficient deployment, making him a key figure in the push toward truly autonomous, continuously learning robots.