Design concept

Raymond Sheh

An unofficial concept study. Enter the passphrase to view.

Research

Research

Most people pick a lane. My work lives in the seams between fields, because that is where the hard, consequential problems hide. The common thread is measurement. If you cannot test it, you cannot trust it.

01

Five fields, one method

01

Measurement Science

Defining the tests that turn "seems fine" into evidence a government can procure against.

02

Responsible AI

Making AI safe, explainable, and defensible in the applications where being wrong is expensive.

03

Robotics

Intelligent machines that work when it matters: disaster zones, factories, operating rooms.

04

Cybersecurity

An intelligent system you cannot secure is a liability, however capable it is.

05

Standards

Turning all of the above into shared rules the field can build and buy against.

02

Selected publications

Two decades of peer-reviewed work on explainable AI and response robotics, cited more than a thousand times. Live metrics and the full list on Google Scholar.

Cited 0 times, and still compounding.
More than a thousand citations across two decades of explainable-AI and response-robotics research, with 432 of them since 2021.
0
Citations
+432 since 2021
0
h-index
13 since 2021
0
i10-index
14 since 2021
Most-cited work: 141 citations for "Effective user interface design for rescue robotics" (2006).
141cited
Effective user interface design for rescue robotics Most citedKadous, Sheh & Sammut · ACM SIGCHI/SIGART Conf. on Human-Robot Interaction · 2006
68cited
"Why did you do that?" Explainable Intelligent RobotsSheh · AAAI-17 Workshop on Human-Aware AI (HAAI-17) · 2017
64cited
Defining explainable AI for requirements analysisSheh & Monteath · KI – Künstliche Intelligenz · 2018
50cited
16 years of RoboCup RescueSheh, Schwertfeger & Visser · KI – Künstliche Intelligenz · 2016
49cited
Robot-assisted laser surgical systemKhan, Fick, Robertson, Sheh, Ironside & Chipper · US Patent 11,369,435 · 2022
46cited
Advancing the state of urban search and rescue robotics through the RoboCupRescue Robot League competitionSheh, Jacoff, Virts et al. · Field and Service Robotics · 2012
44cited
Extracting terrain features from range images for autonomous random stepfield traversalSheh, Kadous, Sammut & Hengst · IEEE Int. Workshop on Safety, Security and Rescue Robotics · 2007
44cited
A low-cost, compact, lightweight 3D range sensorSheh, Jamali, Kadous & Sammut · Australasian Conf. on Robotics and Automation · 2006
03

Two threads I keep pulling

Explainable AI

"Why did you do that?"

Explainability is not a feature you bolt on. It is a requirements problem: different applications need different explanations, and the engineering task is matching the need to what the technique can actually deliver.

To be trusted, a system must explain its decisions and convince us it is making them for the right reasons.
SourceDepthScope
Response Robotics

Measuring the machines we send in first

Standard test methods for the robots that go into rubble, fire, and flood. Developed with DHS and ASTM, hardened at RoboCup Rescue, and pressure-tested in the DARPA Robotics Challenge. If it cannot be scored on a course, it cannot be trusted in a disaster.

Pioneering measurement science to define the gaps, then engaging the research community to close them.
04

By the numbers

30+
standard test methods for response robots
NIST, since 2011
$200M
in US procurements those tests guided
2012–2020
17
peer-reviewed papers, plus book chapters
Career
1
patented robotic laser surgical system
US 11,369,435