Jason Dekarske

Convincing Robots to Trust Humans

RASCAL Lab | HOME STRI | | Blog | Live CV

Currently

Jan 2020- UC Davis
Graduate Student Researcher

As humans wade further from Earth, we must develop technologies to support our survival. The Habitats Optimized for Missions of Exploration (HOME) research institute develops a basic understanding of how habitats interact with humans using mixed autonomy. Robotic systems must collaborate with humans when necessary and maintain themselves when humans are away. I’m studying high-level robotic decision-making in collaborative tasks. See the Education section below for more information about my research.

Education

See Linkedin for courses

2018-2024 University of California, Davis
PhD, Mechanical and Aerospace Engineering

• Advised by Dr. Sanjay Joshi
• Research Highlights:
  3D printed semi-active hand prostheses
  Meta-models of human cognitive states and robots
  Behavioral measures of knowledge, trust, workload, situational awareness
  Deep learning models for time evolving human trust
  Formal planning methods (traditional Artificial Intelligence)
  Full stack ROS systems incorporating UR5e industrial robotic manipulator with remote control GUI
  Remote (web browser based), simulated, and colocated robot interaction experiments
  EMG HMI testing
• ROS robot arm repository
• ROS astrobee repository
• EMG repository

2014-2018 University of Wisconsin, Madison
BSc, Biomedical Engineering

• Advised by Dr. Peter Adamczyk
• Research Highlights:
  Kinematic analysis with motion capture
  Prosthetic device design
  Closed loop impedance controller
• Prosthesis repository

Professional Training

May 2023-Aug 2023 NASA Science Mission Design School
Planetary Science Summer School

With a team of early career scientists and engineers, I developed a conceptual mission to Neptune’s moon, Triton. Planetary Science Summer School is an 11-week program including rigorous preparation for a final week of concurrent mission design with NASA Jet Propulsion Lab’s Team-X. In this experience, I served as Project Manager and organized teamwork, mission requirements, and spacecraft capabilities. Our mission aimed to characterize the subsurface ocean of Triton, the enigmatic surface plumes, and unique cantaloupe terrain. The posters we presented at AGU 2023 can be found on my website. more info…

October 2020 International Conference on Automated Planning and Scheduling
Tutorial on Epistemic Planning and Summer School on Automated Planning & Scheduling

I had the opportunity to learn planning fundamentals like PDDL, Solvers, Execution Monitoring, and Robotics.

Experience

In addition to the professional experience listed below, you can find my infinitely expanding hobby, Open Source, and self-teaching projects on Github. I have a habit of dreaming about an end product and learning the skills needed to produce it, whether or not those skills are in my wheelhouse. So I’m hoping that the software I share can improve community-driven projects and that snippets of thoughts can help other learners.

Aug 2021-June 2022 NASA Ames Research Center
Intern Intelligent Robotics Group

Distant robotic systems operate under significant time delay, posing teleoperation challenges for earthly operators. Predictive systems could fill the temporal void between commands sent to a robotic system. Still, proposed techniques rely on precisely modeled systems. In this project, I’m studying the performance of a human-astrobee team operating under time delay with an idealized predictive system.

Jul 2019-Jul 2021 Ravata Solutions
Consulting Control Systems Engineer

• Designed and built high-speed micro pressure multiplexor
• Performed system identification on microfluidic flow
• Implemented closed-loop controller and estimator for embryo position control
• Implemented system in embedded C++ on custom hardware
• Built GUI using PyQT for touchscreen interface
• Designed microfluidic chip physical interface using 1D tolerance analysis and Fusion360
• Set-up and managed data pipelines with InfluxDB (IC Stack) and Python with on-site server
• Implemented Google Identity Platform for company access of the database
• Mentored interns in mechatronic design fundamentals

2018-2019 UC Davis
TA - Dynamics

• Prepared practice problems to help students gain intuition about how the world moves.
• Led discussion sections working through practice problems.
• Led office hours for individualized teaching sessions.

2018 Ottobock HealthCare
R&D Intern - Mechatronics

• Analyzed gap in commercial and academic active prosthetic foot solutions
• Designed testbed foot for exploration
• Included dynamic variable stiffness actuation and series elasticity
• Optimized design using a Monte Carlo technique
• Conducted kinematic and quasi-static analyses on proposed design
• Sourced parts and submitted technical drawings for fabrication

2016-2018 UW BADGER Lab
Research Assistant

• Set up and support $20,000 motion capture/force plate system using custom MatLab scripts
• Designed and fabricated cutting edge testing apparatus for trans-tibial prostheses
• Developed hardware and software for custom 450mm quadcopter and 6 DOF robot arm for UW Robotics courses
• Designed custom digital filter for processing pressure insole gait trials
• Quadcopter repository

2016 UW Stout
REU Researcher

• Developed power wheelchair computer vision navigation system
• Used Python with Raspberry Pi systems
• Submitted with entirely open-source components
• Wheelchair repository

Skills

Research Research Question Development, Literature Review, Human Experiment Design and Facilitation, IRB Documentation, Publication Preparation, Public Speaking, Science Communication, Cognitive State/Human Performance Assessment using NASA TLX, Multidimensional Trust Scales, Situation Awareness Rating Technique
Control System Synthesis Systems Modeling, Simulation, Matlab, Controller and Estimator Design
Mechatronics ROS/Gazebo, Inventor/Fusion360/SolidWorks, EAGLE, CNC Fab, 3d Printing, Digital Circuit Design, Embedded C/C++, FreeRTOS, STM32Cube
Computing Linux, Bash, Git, Python (PyTorch, NumPy, PyMC, SciPy, SymPy), MongoDB, InfluxDB, Embedded C/C++, Rust, Docker, Networking, Typescript (Javascript, Three.js, Vue), HTML/CSS
Learning ESP-IDF, Flutter, FMCW Doppler Radar Design, Curling, Golf (not scratch yet…), Snowboarding (haven’t landed a backflip… on purpose)
Languages English (Native), German (B1)

Outreach

Adaptive Adventures Help people with disabilities rock climb
National Biomechanics Day Taught interactive lessons to all ages
MESA Python Workshop Taught Python fundamentals to high schoolers from historically under represented ethnic groups
FIRST Robotics Judge competitions based on students’ robotics knowledge and implementation
ENG 003 Design Judge undergraduate student engineering projects

Awards

2019 UCD Graduate Research Award

2018 UC Davis MAE Departmental Fellowship
Tech B2B Innovation Program 3rd Place
Tong BME Design Award 2nd Place

2016 Kohler Foundation Incentive Scholarship
Denn-Kroeff Scholarship

Presentations

2023 ACM Conference on Computer-Supported Cooperative Work and Social Computing
Workshop Chair
Emerging Telepresence Technologies for Hybrid Meetings: an Interactive Workshop

2020 Habitats Optimized For Missions of Exploration Technical Seminar
Architecture for a Distributable, Remote Human-Robot Interaction Experiment

2016 Biomedical Engineering Society Annual Meeting
Path Oriented Powered Wheelchair Navigation Assistance (Poster)

Publications

2024 Jason Dekarske, Zhaodan Kong, Sanjay Joshi Dynamic Human Trust Modeling of Autonomous Agents With Varying Capability and Strategy, Human Factors: The Journal of the Human Factors and Ergonomics Society, in review.

2023 Ayush Mohanty, Jason Dekarske, Sanjay Joshi, Nagi Gabraeel, Operational Self Awareness of Robotic Manipulators Performing Tasks with Dynamic Severity Levels, IEEE Transactions on Systems, Man and Cybernetics: Systems, in review.

Gregory Bales, Allison Anderson, Jason Dekarske, Sanjay Joshi and Zhaodan Kong, An Investigation of EEG-Network-based Dynamics Trust Metrics in Human-autonomy teaming, IEEE Transactions on Human Machine Systems, in review.

Amanda Steckel, …, Jason Dekarske, …, The Science Case of Nautilus: A Multi-flyby Mission Concept to Triton. American Geophysical Union Annual Meeting 2023, poster.

Tyler Horvath, …, Jason Dekarske, …, The Architecture of Nautilus: A Multi-flyby Mission Concept to Triton. American Geophysical Union Annual Meeting 2023, poster.

2022 Jason Dekarske, Zhaodan Kong and Sanjay Joshi, Operationalizing Human Cognitive State Assessment (poster), SmartHab Workshop, San Antonio, TX.

2021 J. Dekarske and S. S. Joshi, “Human Trust of Autonomous Agent Varies With Strategy and Capability in Collaborative Grid Search Task,” 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS), 2021, pp. 1-6, doi: https://doi.org/10.1109/ICHMS53169.2021.9582622.

2020 Mrazsko, H. X., Dekarske, J. T., and Adamczyk, P. G. (June 26, 2020). “Design of a Variable Stiffness Pneumatic Ankle Prosthesis With Self-Recharging for Weightlifting Exercise.” ASME. J. Med. Devices. September 2020; 14(3): 034501. https://doi.org/10.1115/1.4047493