My role as Senior Director comprises two main areas of responsibility:- I oversee the Robotics Software Team. I strongly believe in letting talented people do what they do best, and so I empower my direct reports to run their groups as they see fit, providing guidance as needed.- I run the Locus Enhanced Autonomy Program (LEAP), which is an R&D program aimed at bringing novel solutions from academia into our software stack.

After my promotion to Director, I continued to be involved with our solution technically:- Improved robot bag (log file) recording and playback to aid in analysis- Implemented nav-in-the-loop validation tool for site configurations- Increased the robustness of our docking algorithm- Continued to improve our lightweight communication stack- Increased the robustness of our navigation stack- Wrote extensive documentation for our navigation stack to enable junior engineers to be onboarded rapidlyIn addition, I was responsible for overall management of the Robotics Software Team:- Successfully implemented new title ladder- Formalized previously unofficial group structure and promoted engineers to group leadership positions- Interviewed numerous candidates for open positions- Doubled the size of the software team- Worked with product management to increase formalization of our development processes- Helped to coordinate the work required to support our new Origin II platform- Delivered numerous features on time

In my role as Principal Roboticist, I continued working on many areas of our software stack:- Implemented novel approach to secondary safety control loop for collision prevention- Implemented "anytime" path tracking approach to ensure stability in navigation under high CPU load- Implemented a regression testing package for state estimation- Implemented extremely lightweight communication protocol for sharing of robot data- Implemented new navigation recovery behaviors and improved others- Ran our navigation group: sprint planning, managing deliverables, etc.

As a Senior Roboticist for Locus Robotics, I worked on a wide array of problems:- Improved the accuracy of state estimation through improved filtering and calibration- Wrote a highly efficient laser scan-based robot localization module using state-of-the-art algorithms from robotics literature- Enabled our robot to repeatably and reliably dock autonomously with its charging station. This involved novel approaches to high-accuracy localization and control.- Was instrumental in the development of Locus Robotics Advanced Navigation (LRAN) technology. This involved both improving existing codebases and writing new modules.- Made critical improvements to our Simultaneous Localization and Mapping software- Deployed an automated build server to facilitate rapid deployment of binary packages to robots across customer sites- Actively supported customer sites- Acted as release manager for each release of our robot software stack. This included tracking (and often fixing) regressions, coordinating with team members, and involving senior management in the approval process.- Worked with team members to improve our software development processes and standardsMany of the algorithms that I have developed or improved during my time with Locus have resulted in patent applications. To date, all software has been written in C++ or Python, and typically for the Robot Operating System (ROS).I am a remote employee with Locus Robotics, and carry out my tasking from Edinburgh, United Kingdom.

During my tenure as an Autonomy Development Contractor for Clearpath Robotics, I worked on several key components of Clearpath's autonomy framework. My contributions included:- Wrote a specialized path planner for use on various industrial robots- Helped improve both accuracy and stability of navigation framework- Helped to increase the robustness of planning framework, allowing continuous operation in complex environments- Aided in improving Clearpath's fork of the robot_localization ROS packageAll software was based on the Robot Operation System (ROS) framework, and was written in C++. All work, including live robot testing, was performed remotely from the Boston area.

In my role as Software Engineer within the Sensor Processing and Networking division of Charles River Analytics, I have worked on a broad range of projects in autonomous robotics, including:- Autonomous communications relays for long-range unmanned ground vehicle teleoperation- Quadrotor-based visual anomaly detection in crops- Leader following behavior for unmanned ground and aerial vehicles- State estimation for unmanned underwater vehicles- Simultaneous localization and mapping for unmanned underwater vehiclesThe majority of these projects relied on the Robot Operating System (ROS), and were written in C++ for Linux operating systems. Our need for robust state estimation using a large number of sensor inputs led to the creation of the robot_localization ROS package, of which I am the author and maintainer. The package provides state estimation in the form of an extended Kalman filter (EKF) implementation and an unscented Kalman filter (UKF) implementation. It has since been released as open source and is used on thousands of robots around the world.Other accomplishments during my tenure at CRA include:- Porting Charles River Analytics' VisionKit™ software to iOS- Numerous improvements to VisionKit- Applying VisionKit's Object Detection Framework to new problems in the underwater domain

While at SeeByte, I worked almost exclusively on the Autonomous Inspection Vehicle (AIV) project. My contributions to the project included:- Maintained and improved the vehicle's state estimation module- Implemented an underwater localization module so as to extend mission lifetime- Made improvements common software toolkits- Participated in several multi-week trialsSoftware was written in C++ for Linux operating systems. I worked heavily with extended Kalman filters (EKFs), particle filters, and sonar data processing.

In my role as robotics research scientist at Lockheed Martin Advanced Technology Labs, I worked on a broad range of cutting-edge research projects in autonomous robotics. These projects involved:- Covert navigation for unmanned ground vehicles- Dense crowd navigation for unmanned ground vehicles- Navigation and perception for a rotating mono-wing unmanned aerial vehicle- Navigation for large unmanned underwater vehicles- Implicit coordination between robots and humans- Research into robot cognition and learning of affordances- Navigation of a crawling millirobot through complex electronic environmentsMy typical roles on projects included research, software design, software implementation, and testing. Software was written in C/C++, typically for Linux-based operating systems. I implemented numerous algorithms, including A*/D*, Field D*, RRTs, lattice planners, potential fields, image segmentation, and particle filters.

Handled all in-house software development, typically to support data management for a legacy ERP software package. Aided in development and support of IT infrastructure.

Worked with a small group of developers on the Electronic Grants Management Information System (EGMIS) ASP.NET web application for the Pennsylvania Commision on Crime and Deliquency. Tasks included development of new functionality, testing, and bug fixing.

Aided in development and support of IT infrastructure.