M. Herbert

About

M. Herbert is a pioneering roboticist whose work has fundamentally shaped autonomous vehicle and mobile robot perception. Best known for leading the landmark CMU Navlab project, Herbert developed the core perception systems enabling autonomous driving in challenging real-world conditions—including bad weather, poor lighting, and degraded roads. Their 1991 paper on Navlab perception, with 117 citations, introduced road-following techniques using color classification and neural networks that became foundational for the field. A companion paper on the Navlab's EDDIE architecture (47 citations) provided a toolkit for building mobile robot systems, influencing generations of autonomous vehicle design. Herbert's research spans active learning for outdoor perception, sensor-based interior modeling for nuclear facility decommissioning, and brain-machine interface control of robotic arms enhanced by computer vision. Their 2006 work on active learning addressed the critical challenge of adapting perception systems to new environments without extensive hand-tuning. Herbert's contributions bridge theoretical machine learning and practical robotics, demonstrating how intelligent perception systems can operate robustly across diverse domains—from autonomous cars to assistive neuroprosthetics—making them a key figure in bringing robots from laboratories into the real world.

Research Focus

Key Achievements

Top Papers

1. 1
   Toward autonomous driving: the CMU Navlab. I. Perception

   117 citations · 1991
2. 2
   Toward autonomous driving: the CMU Navlab. II. Architecture and systems

   47 citations · 1991
3. 3
   Sensor-based interior modeling

   10 citations · 1995
4. 4
   Active learning for outdoor perception

   7 citations · 2006
5. 5
   201 Brain-Machine Interface Control of a Robotic Arm for Object Grasping is Improved With Computer-Vision Based Shared Control

   3 citations · 2015

Key Collaborators

Contact & Links