Jason Wakim Richards
I am an aerospace engineer and doctoral researcher working in guidance, navigation and control (GNC). I connect estimation, simulation, and deployed robotics - from filter design and physics models to lab and field testing.
Currently a doctoral researcher at SnT, University of Luxembourg, working on Semantic Relational SLAM for UAVs with ROS 2 and PX4.
Engineering approach
My work connects sensor data, state estimation, and control with the vehicle and its operating environment.
Featured work
A ROS 2 and Gazebo simulation where I develop stereo visual odometry, wheel odometry and a continuous error-state filter around a six-wheel rover.
Featured project
ROS 2 Jazzy and Gazebo Harmonic stack in C++17 with Eigen and OpenCV. I develop the odometry front ends, the error-state filter, Ackermann control and the rosbag plus Plotly review path. More information on scope and limits is on the projects page.
Simulation and estimation
Two compact codebases: an entry-vehicle trajectory simulator and a rigid-body and gravity framework with a CubeSat ADCS harness.
A C90 and CMake physics core with a layered Earth atmosphere and six-degree-of-freedom trajectory integration. I implemented the atmosphere, RK4 and Dormand-Prince 5(4) solvers, and archive loop, with Python tooling to inspect the atmosphere model.
I developed NBodySimulator to simulate and test JamSail's ADCS: an Extended Kalman Filter fusing sun-sensor, magnetometer and IMU measurements, my own IGRF implementation, and a sun-sensing algorithm developed and tested in the simulator.
Toolbox
Now and recently
Full history lives on the Background page and the printable CV.
Semantic Relational SLAM for UAVs. ROS 2 and PX4 development in C++ and Python with OpenVINS, Docker, Kalibr and embedded Linux, across simulation, lab and flight testing.
Led ADCS requirements, interfaces and embedded C development. Used NBodySimulator to test EKF sensor fusion and sun sensing, with an IGRF model I implemented.
LiDAR integration and testing, including test-bench and breadboard campaigns, plus C/C++ rover simulation with synthetic camera and LiDAR data.
For questions about the simulators, ADCS work or doctoral research, the best route is professional email. I also keep public code online.