Papers
11
Total Citations
848
H-Index
8
About
Douglas Morrison is a leading roboticist whose work sits at the intersection of robotic grasping, manipulation, and semantic perception. He is best known for developing the Generative Grasping Convolutional Neural Network (GG-CNN), a breakthrough approach that enables real-time, closed-loop grasp synthesis directly from depth images. This work, detailed in his highly cited 2019 paper (474 citations), overcomes the limitations of traditional grasp planning by predicting pixel-wise grasp quality and pose, allowing robots to reactively adjust grasps in dynamic environments. Morrison’s impact extends to semantic understanding in robotics, with his comprehensive survey on semantics for mapping, perception, and interaction (100 citations) providing a foundational reference for the field. He also led the perception and grasping system for Cartman, the low-cost Cartesian manipulator that won first place at the 2017 Amazon Robotics Challenge, demonstrating robust performance in cluttered scenes with limited training data. His contributions have advanced both the theory and practice of robotic manipulation, making real-time, object-independent grasping a tangible reality for industrial and service robotics.
Research Focus
Key Achievements
Top Papers
- 1Learning robust, real-time, reactive robotic grasping474 citations · 2019
- 2
- 3Semantics for Robotic Mapping, Perception and Interaction: A Survey100 citations · 2020
- 4Semantic Segmentation from Limited Training Data52 citations · 2018
- 5Semantics for Robotic Mapping, Perception and Interaction: A Survey44 citations · 2020
- 6
- 7
- 8Mechanical Design of a Cartesian Manipulator for Warehouse Pick and Place10 citations · 2017
- 9
- 10Semantic Segmentation from Limited Training Data5 citations · 2017