John J. Leonard
John Leonard is an MIT roboticist who studies how autonomous robots navigate and map their surroundings. He earned engineering degrees at the University of Pennsylvania and the University of Oxford, coauthored a 1992 robot-mapping paper, and advised MIT students building an autonomous underwater robot in 1997. He led MIT’s team in the 2007 DARPA Urban Challenge, became an IEEE Fellow in 2014, and joined Toyota Research Institute’s technical team for autonomous driving in 2016. The AAAS elected him a Fellow in 2019, and his recent work includes underwater scene reconstruction, language-model-assisted object SLAM, and diver-AUV teaming. As of October 2026, he is an MIT mechanical engineering professor and Associate Department Head for Education.
- Works
- 318
- Citations
- 26,174
- h-index
- 76
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
1987–1994
Degrees and robot mapping
He earns a B.S.E.E. in electrical engineering and science from the University of Pennsylvania in 1987 and a D.Phil. in engineering science from the University of Oxford in 1994. He coauthors the 1992 paper “Dynamic map building for an autonomous mobile robot” with Hugh F. Durrant-Whyte and Ingemar J. Cox.
-
1997
Advising an underwater robot
In 1997, he is an assistant professor of marine robotics at MIT and advises ocean engineering students building Autolycus, a robot they hope will swim underwater on its own.
-
2007
Leading MIT’s robocar
In 2007, he leads MIT’s team in the DARPA Urban Challenge and takes its self-driving vehicle to the semifinals.
-
2014–2016
IEEE Fellow and Toyota team member
He becomes an IEEE Fellow in 2014. In 2016, he joins Toyota Research Institute’s technical team for autonomous driving and works part time while retaining his MIT role.
-
2019
An AAAS fellowship
In 2019, the AAAS elects him a Fellow. His research focuses on long-term visual SLAM in dynamic environments and self-driving cars.
-
2025
SeaSplat and language-model SLAM
In 2025, he coauthors SeaSplat, a study on 3D Gaussian splatting for underwater scenes and color correction. He also coauthors an IROS paper titled “Semantic Enhancement for Object SLAM with Heterogeneous Multimodal Large Language Model Agents.”
-
2026
Diver-AUV teaming and ReefMapGS
By 2026, his Marine Robotics Group’s diver-AUV teaming work includes simulations and calm-water field tests with human-paddled kayaks as proxies for divers and AUVs. In April 2026, he coauthors ReefMapGS, a preprint on incremental underwater scene reconstruction with multimodal SLAM and 3D Gaussian splatting.
-
As of October 2026
Associate head for education
As of October 2026, he is Samuel C. Collins Professor of Mechanical Engineering and Associate Department Head for Education at MIT.
Sources and credits
Photo: We have no published photo of John J. Leonard yet. Send us a free-licensed photo.
- John Leonard | MIT CSAIL
- Dynamic map building for an autonomous mobile robot - University of Copenhagen Research Portal
- Students get feet wet during lab-research workshop | MIT News | Massachusetts Institute of Technology
- DARPA names MIT's 'robocar' a semifinalist | MIT News | Massachusetts Institute of Technology
- Toyota AI Team Hires James Kuffner from Google Robotics, Will Have Rodney Brooks as Adviser - IEEE Spectrum
- Six MIT faculty elected 2019 AAAS Fellows | MIT News | Massachusetts Institute of Technology
- Imaging technique removes the effect of water in underwater scenes | MIT News | Massachusetts Institute of Technology
- Publications - Marine Robotics Group at MIT
- Human-machine teaming dives underwater | MIT News | Massachusetts Institute of Technology
- [2604.11992] ReefMapGS: Enabling Large-Scale Underwater Reconstruction by Closing the Loop Between Multimodal SLAM and Gaussian Splatting
- MECHE PEOPLE: John Leonard | MIT Department of Mechanical Engineering