Adam Coates

Papers

Total Citations

272

H-Index

About

Adam Coates is a leading researcher in robotics and machine learning, whose work has fundamentally advanced how machines perceive and interact with the world. His early research tackled the foundational challenge of state estimation, developing "Discriminative Training of Kalman Filters" (123 citations) to automate the tuning of critical modeling parameters, a problem that previously required extensive manual effort. Coates then pioneered the use of GPU hardware for "Scalable learning for object detection" (52 citations), dramatically accelerating complex algorithms and making real-time robotic vision practical. He further enhanced perception systems by developing probabilistic methods for "Multi-camera object detection" (51 citations), enabling robots to robustly identify objects from multiple viewpoints in cluttered environments like homes and offices. A particularly notable achievement is his work on "Autonomous sign reading for semantic mapping" (46 citations), where he integrated text detection and recognition into SLAM systems, allowing robots to automatically annotate maps with meaningful semantic labels. Through these contributions, Coates has been instrumental in bridging the gap between raw sensor data and actionable, semantic understanding, laying the groundwork for more intelligent and autonomous robotic systems.