Jonathan Claassens

Council for Scientific and Industrial Research, Council of Scientific and Industrial Research

Papers

Total Citations

H-Index

About

Jonathan Claassens is a roboticist whose research bridges the gap between human demonstration and autonomous robot execution, with a particular focus on path planning, object recognition, and localization in challenging environments. His most influential work, "An RRT-based path planner for use in trajectory imitation" (2010, 13 citations), introduces a robust programming-by-demonstration system that uses Rapidly-exploring Random Trees (RRT) to generate reproduction paths satisfying statistical constraints from human demonstrations while avoiding obstacles. This contribution enables robots to learn and replicate complex tasks with greater flexibility and safety. Claassens further advanced active perception with his work on "Active object recognition using vocabulary trees" (2013, 8 citations), which improves mobile robot object classification accuracy by intelligently changing viewpoints. His investigation into trajectory estimation for underground mining (2014, 4 citations) addresses the critical challenge of robot localization in GPS-denied, harsh environments using time-of-flight cameras and inertial measurement units. Additionally, his exploration of affordance symmetries for task reproduction (2012, 3 citations) provides a novel framework for robots to exploit inherent task freedoms—such as cutting a carrot anywhere along a knife blade—enabling more natural and efficient tool use. Claassens’s work has been cited over 28 times, reflecting its impact on advancing autonomous robotics in both structured and unstructured settings.