Jeffrey Kane Johnson

Robotics & Computer Science

About

I work to develop safety-verifiable, vision-based methods for autonomous robotic collision avoidance. In general my research interests revolve around perception-based motion planning and control for use in multi-agent systems.

Projects

Education

PhD in Computer Science, 2012—2017
School of Informatics, Computing, and Engineering
Indiana University, Bloomington, Indiana
IU Computer Vision Lab
MSc in Computer Science, 2009—2012
School of Informatics, Computing, and Engineering
Indiana University, Bloomington, Indiana
Zertifikat Deutsch (CEFR level B1), 2004
VHS Recklinghausen, Germany
BSc in Computer Science, 1999—2003
Trine University, Angola, Indiana

Professional

Mapless AI, Inc., March. 2020—Current
Principal
- Research in applied artificial intelligence and autonomous system safety.
Maeve Automation, Feb. 2019—March 2020
Researcher
- Advanced, autonomy-based vehicle driver assistance systems
Uber ATG, Jan. 2018—Feb. 2019
Senior Autonomy Engineer
- Motion planning for urban automated vehicles
Apple, Inc., Jan. 2016—May 2017
Engineer
- Experimental algorithm and software development for autonomous systems
Robert Bosch, LLC, Jan. 2014—Nov. 2015
Research Engineer
- Lead development of motion planning/decision making for automated driving
Robert Bosch, LLC, May 2013—Aug. 2013
Intern
- Collision detection methods for optimization-based vehicle motion planning
TRACLabs, June 2012—Aug. 2012
Intern
- Software toolkits for coordinated dual-arm manipulation
Demo Videos
Indiana University, Sept. 2006—Sept. 2009
Web Developer, School of Journalism
- Development of school web site and internal web-based applications
Software Developer, March 2004—Sept. 2006
Freelance
- Software development for small to mid-sized clients

Affiliations

IEEE Intelligent Vehicles Symposium - IV2020
Associate Editor
Workshop on Ensuring and Validating Safety for Automated Vehicles (IV2020)
Co-proposer
US Robotics Roadmap Workshop, Aug. 2019—Current
Member
STP: UL 4600 Standard for Safety for the Evaluation of Autonomous Products, July 2019—Current
Stakeholder
IEEE ITSS Technical Committee on Self Driving Automobiles, Nov. 2018—Current
Member
- Co-chair Cooperative Interacting Vehicle Workshop in IEEE Intelligent Vehicles Symposium - IV2019

Publications

The Colliding Reciprocal Dance Problem: A Mitigation Strategy with Application to Automotive Active Safety Systems
American Control Conference (ACC) 2020
Jeffrey Kane Johnson
(pdf) (bib)
Safe Motion Planning under Partial Observability with an Optimal Deterministic Planner
American Control Conference (ACC) 2020
Jeffrey Kane Johnson
(pdf) (bib)
Visual Servoing for Mobile Ground Navigation
IEEE Connected and Automated Vehicles Symposium (CAVS) 2018
Jeffrey Kane Johnson
(pdf) (poster) (bib)
On the Relationship Between Dynamics and Complexity in Multi-agent Collision Avoidance
Autonomous Robots (AURO)
Jeffrey Kane Johnson
(pre-print pdf) (published version) (bib)
Guest talk: Vision-based Navigation for Autonomous Vehicles
Intelligent & Interactive Systems Talk Series
Jeffrey Kane Johnson
(Video) (Slides)
Image Space Potential Fields: Constant Size Environment Representation for Vision-based Subsumption Control Architectures
Maeve Automation Technical Report 1
Jeffrey Kane Johnson
(pdf) (bib)
Encroachment Detection with Monocular Vision for Small, Low-cost, Compute-constrained Platforms
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2017
Abstract Only
Jeffrey Kane Johnson
(pdf) (poster) (bibtex)
Selective Determinism for Autonomous Navigation in Multi-agent Systems
Ph.D. Dissertation
Jeffrey Kane Johnson
(pdf) (bibtex) (slides)
Constant Space Complexity Environment Representation for Vision-based Navigation
IROS 2017 9th Workshop on Planning, Perception and Navigation for Intelligent Vehicles
Jeffrey Kane Johnson
(pdf) (poster) (slides) (bibtex)
A Novel Relationship Between Dynamics and Complexity in Multi-agent Collision Avoidance
Robotics: Science and Systems (RSS) 2016
Jeffrey Kane Johnson
(pdf) (slides) (poster) (bibtex)
Identifying Support Surfaces of Climbable Structures from 3D Point Clouds
IEEE International Conference on Robotics and Automation (ICRA) 2014
Anna Eilering, Victor Yap, Jeff Johnson, Kris Hauser
(pdf) (bibtex)
Optimal Longitudinal Control Planning with Moving Obstacles
IEEE Intelligent Vehicles Symposium (IV) 2013
Jeff Johnson, Kris Hauser
(pdf) (web) (slides) (bibtex)
Optimal Longitudinal Control Planning with Moving Obstacles
ICRA 2013 Workshop: Vehicle Autonomy in Urban Transportation Systems
Jeff Johnson, Kris Hauser
(pdf) (bibtex)
Minimizing Driver Interference Under a Probabilistic Safety Constraint in Emergency Collision Avoidance Systems
IEEE Intelligent Transportation Systems Conference (ITSC) 2012
Jeff Johnson, Yajia Zhang, Kris Hauser
(pdf) (slides) (bibtex)
Optimal Acceleration-Bounded Trajectory Planning in Dynamic Environments Along a Specified Path
IEEE International Conference on Robotics and Automation (ICRA) 2012
Jeff Johnson, Kris Hauser
(pdf) (slides) (bibtex)
Semiautonomous Longitudinal Collision Avoidance Using a Probabilistic Decision Threshold
IROS 2011 International workshop on Perception and Navigation for Autonomous Vehicles in Human Environment
Jeff Johnson, Yajia Zhang, Kris Hauser
(pdf) (bibtex)

Experiment Recordings

These data sets are recordings of closed-loop experiment runs that demonstrate library functionality. They are recorded as rosbags. All data sets released under CC BY 4.0 license.

e02: Longitudinal Colliding Reciprocal Dance
e02: Longitudinal Colliding Reciprocal Dance
  • Demonstration of colliding reciprocal dance issue in simple longitudinal scenario
  • Three mitigation strategies are testing: constraint tightening, conservative deceleration, and none
  • Includes RViz config for easy viewing
e01: Short AR slalom with Image Space Potential Field Navigation
e01: Short AR slalom with Image Space Potential Field Navigation
  • Six second run demostrating ar_isp_field control package
  • Guidance control is constant forward throttle and centered steering
  • Staggered obstacles obstruct the agent's path
  • The controller computes ISP fields from camera input and perturbs the guidance control to avoid collision
  • Slideshow for experiment: https://youtu.be/BaHekhZmkfY

Simulation Sequences

These data sets are open-loop simulation sequences provided by Parallel Domain. All data sets released under CC BY 4.0 license.

s05: Hills with transverse traffic
s05: Hills with transverse traffic
  • 175 frame, medium resolution visual of traveling along a hilly road with transverse traffic
  • Ground truth, per-pixel semantic segmentation with blur noise
  • 1280x720 resolution
  • Includes source simulation movies
s04: Road fork
  • 300 frame, medium resolution visual of an approach and traversal of a forked road
  • Ground truth, per-pixel semantic segmentation with blur noise
  • 1280x720 resolution
  • Includes source simulation movies