Jonathan Spraggett
Robotics Software Engineer @Mostavio
Signup · Get unlimited contacts
WORK HISTORY
Robotics Software Engineer @Mostavio
Toronto, ON, CA
Lead the development of autonomy, motion-planning, and state estimation software for a 250 kg eVTOL aircraft using ROS2, C++ and Python. Implemented sensor fusion (Stereo Cameras, GPS, UWB & LiDAR) with residual-based outlier rejection, improving localization reliability under GPS‑denied conditions.• Developed real-time trajectory planners (Pure Pursuit, DWA, A*) and optimized flight controlloops, reducing tracking RMSE by 63% and improving autonomous-flight stability. Validatedthe system in flight tests, demonstrating reliable obstacle avoidance and fail-safe behaviour.• Built high‑fidelity SITL/HITL Gazebo simulations with integrated CI/CD pipelines with Docker,cutting regression‑test cycle from hours to <20 min. Codebase now ships as container images forseamless field deployment.• Contributed to Mostavio winning GoAERO Stage 1 for autonomous eVTOL innovation (2025).
EDUCATION
University of Toronto
Bachelor of Applied Science in Engineering Science - BASc , Robotics Engineering
ABOUT JONATHAN SPRAGGETT
I build autonomy software that turns cutting-edge research into robots you can trust. Right now, I lead robotics software R&D at Mostavio, guiding a 250 kg eVTOL from simulation (Gazebo SITL/HITL) to real-world test flights. Early in the week, I’m in the lab writing high-rate ROS 2 sensor drivers in C++/Python on a Jetson Orin and squeezing centimetre-level accuracy out of residual-based Kalman filters. By week’s end, I’m building GPU-accelerated perception pipelines and flight-testing the eVTOL. Recently, I redesigned a core component of the control system with digital filters for vibration mitigation. Those efforts helped Mostavio win the Stage 1 GoAero Prize.During undergrad, I led the development of a fully autonomous humanoid soccer robot, finishing 4th globally at RoboCup 2023 in Bordeaux. I am still advising the team and enjoy mentoring junior engineers. This project sparked my first-author paper, “Learning to Get Up Across Morphologies: Zero-Shot Recovery with a Unified Humanoid Policy” (RoboCup Symposium 2025), and cemented my fascination with deep-RL control on custom hardware.I hold a BASc in Engineering Science (Robotics major, AI minor) from the University of Toronto and continue to publish on Reinforcement Learning, eVTOL Control Systems, and Safety-Critical Autonomy. I thrive where rigorous research meets real-world deployment and welcome collaboration on frontier robotics challenges.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.