William D. Richards
Papers
1
Total Citations
5
H-Index
1
About
William D. Richards is a roboticist whose research lies at the intersection of computer vision and autonomous manipulation, with a particular focus on enabling robots to perceive and interact with challenging, real-world environments. His most notable contribution, "Depth by Poking: Learning to Estimate Depth from Self-Supervised Grasping," addresses a critical limitation in robotic perception: the failure of standard depth sensors on reflective or transparent surfaces. By training a neural network to infer depth through physical interaction—literally "poking" objects—Richards pioneered a self-supervised approach that allows robots to learn depth estimation from their own grasping attempts. This work, which has garnered 5 citations, exemplifies his commitment to closing the perception-action loop. Richards’ research is distinguished by its practical, data-driven methodology, leveraging real-world robotic experience rather than synthetic datasets. His contributions are particularly impactful for industrial and service robotics, where reliable manipulation of diverse materials is essential. Through this innovative fusion of active sensing and deep learning, Richards is helping to build robots that can see and handle the world as it truly is, not just as sensors imagine it.
Research Focus
Key Achievements
Top Papers
- 1