Andrea Oddera

Papers

Total Citations

H-Index

About

Andrea Oddera's research lies at the intersection of computer vision and robotic manipulation, with a particular focus on enabling robots to interact with their environments through visual feedback. Her most influential work, "A Vision-Based Learning Method for Pushing Manipulation" (1993, 44 citations), introduced an unsupervised, on-line learning approach that allows a robot to push an object connected via a rotational point contact to a desired location in image-space—a foundational contribution to vision-guided manipulation. In a related paper, "A Direct Approach to Vision Guided Manipulation" (1993, 6 citations), Oddera advanced the field by proposing a method that leverages direct image-space calculations of optical flow for continuous real-time control, deriving state variables from optical flow measurements to guide manipulative actions. This work is notable for its departure from traditional model-based approaches, emphasizing real-time adaptability and visual servoing. Though her publication record is concise, Oddera's contributions have been recognized as early and innovative steps toward integrating perception and action in robotics, inspiring subsequent research in learning-based manipulation and visual feedback control. Her work remains a touchstone for those exploring how robots can learn from visual observation to perform complex physical tasks.