Pomerleau

About

Dean Pomerleau is a pioneering figure in autonomous vehicle technology and neural network computing. His research centers on real-time neural network architectures for robotic navigation, particularly the ALVINN (Autonomous Land Vehicle In a Neural Network) system, which demonstrated end-to-end learning for self-driving cars. His seminal 1988 paper, "Neural network simulation at Warp speed," with 161 citations, introduced a breakthrough algorithm that achieved 17 million neural network connections per second using the Warp systolic array computer—a ten-processor parallel system. This work revolutionized the speed of back-propagation training, enabling real-time learning for complex tasks. Pomerleau's contributions laid the foundation for modern autonomous driving systems, showing how neural networks could directly map sensor inputs to steering commands without explicit programming. His achievements include pioneering the use of neural networks for vehicle control and demonstrating the feasibility of learning-based autonomy on public roads. His research continues to influence fields from robotics to machine learning, with his citation impact reflecting decades of foundational work in artificial intelligence and autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Neural network simulation at Warp speed: how we got 17 million connections per second

   161 citations · 1988

Key Collaborators

Contact & Links