Phil Osteen
Mechanical Engineer at US Army Research Laboratory
Baltimore, Maryland, United States
Summary
👤
Senior🎓
Top SchoolPhil Osteen is a mechanical engineer with 12 years of experience applying sensor fusion and algorithm development to improve robot autonomy, currently researching environment-description methods at the US Army Research Laboratory. He has a strong blend of academic and field experience, from contributing to DARPA Urban Challenge systems to building UAV landing simulators and designing perception algorithms at Motile Robotics. Phil’s work spans practical Kalman-filter fusion of GPS/IMU/camera data to novel, correspondence-free 6DoF data association using Spherical Harmonic Analysis for egomotion estimation. He contributes to open-source tooling for visual-inertial calibration (notably enhancements to Kalibr’s multi-IMU and IMU-to-body transform handling), reflecting attention to robust backend sensor transforms. Based in Baltimore, he combines hands-on hardware integration with algorithmic rigor to make robots better understand and interact with real-world, previously unseen environments. Colleagues describe him as a pragmatic researcher who translates complex sensor math into deployable autonomy tools.
12 years of coding experience
7 years of employment as a software developer
Master of Science (MSc), Mechanical Engineering, Master of Science (MSc), Mechanical Engineering at University of Florida
English, Spanish
Github Skills (6)
imu10
calibration10
python10
transformation9
numpy8
camera7
Programming languages (7)
Github contributions (5)
ethz-asl/kalibr
May 2022 - May 2022
The Kalibr visual-inertial calibration toolbox
Role in this project:
Back-end Developer
Back-end DeveloperContributions:5 commits, 1 PR, 3 comments in 4 days
Contributions summary:Phil primarily contributed to the `IccSensors.py` file within the Kalibr toolbox. Their commits focused on modifying the transformation calculations and outputs related to the IMU, including adding the body-to-IMU transformation (T_i_b) to the output and integrating the transformation call. They also added an `imuNr` parameter to enumerate IMUs, indicating work on multi-IMU setups, and updated the method for getting the transformation.
calibration-toolboximutoolboxcameravisual-inertial
phil0stine/image_pipeline
Aug 2019 - Dec 2023
An image processing pipeline for ROS.
Contributions:3 pushes, 3 branches in 4 years 4 months
processing-pipelineimage-processingrospipeline
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.