The human should be part of the control loop?

The capabilities of autonomy have grown to encompass new application spaces that until recently were considered exclusive to humans. In the past, automation has focused on applications where it was preferable to completely replace the human. Today, though, we have the opportunity to leverage the complementary strengths of both human and autonomy technologies to maximize performance and limit risk, and the human should therefore remain “in” or “on” the loop. To adequately assess when and how to accomplish this, it requires us to assess not only the capabilities, but the risks and the ethical questions; coupled to this are the issues with degradation of performance in specific instances (for instance, recovery from failure) that may require a human to remain the sole control authority. This paper investigates the contributors to success/failure in current human-autonomy integration frameworks, and proposes guidelines for safe and resilient use of humans and autonomy with regard to performance, consequence, and the stability of human-machine switching. Key to our proposed approach are (i) the relative error rate between the human and autonomy and (ii) the consequence of possible events.

[1]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[2]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[3]  Philip H. Marsden,et al.  Allocation of functions , 2004 .

[4]  John M. Flach,et al.  Control Theory for Humans: Quantitative Approaches To Modeling Performance , 2002 .

[5]  Birsen Donmez,et al.  Modeling Workload Impact in Multiple Unmanned Vehicle Supervisory Control , 2010, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[6]  S. Shankar Sastry,et al.  Dynamic inverse models in human-cyber-physical systems , 2016, Defense + Security.

[7]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[8]  R. Edlich,et al.  Update on the National Vaccine Injury Compensation Program. , 2007, The Journal of emergency medicine.

[9]  Heesung Kwon,et al.  Human-autonomy sensor fusion for rapid object detection , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[10]  Katja Kircher,et al.  Tactical Driving Behavior With Different Levels of Automation , 2014, IEEE Transactions on Intelligent Transportation Systems.

[11]  Hao Ding,et al.  Structured collaborative behavior of industrial robots in mixed human-robot environments , 2013, 2013 IEEE International Conference on Automation Science and Engineering (CASE).

[12]  Raja Parasuraman,et al.  Complacency and Bias in Human Use of Automation: An Attentional Integration , 2010, Hum. Factors.

[13]  S S Stevens,et al.  HUMAN ENGINEERING FOR AN EFFECTIVE AIR-NAVIGATION AND TRAFFIC-CONTROL SYSTEM, AND APPENDIXES 1 THRU 3 , 1951 .

[14]  Brent Lance,et al.  Confidence Metrics Improve Human-Autonomy Integration , 2014, 2014 9th ACM/IEEE International Conference on Human-Robot Interaction (HRI).

[15]  Wright-Patterson Afb,et al.  F-22 control law development and flying qualities , 1996 .

[16]  Jean-Michel Hoc,et al.  Towards a cognitive approach to human-machine cooperation in dynamic situations , 2001, Int. J. Hum. Comput. Stud..

[17]  Maarten Sierhuis,et al.  Human-agent-robot teamwork , 2012, 2012 7th ACM/IEEE International Conference on Human-Robot Interaction (HRI).

[18]  Allison M. Okamura,et al.  Methods for haptic feedback in teleoperated robot-assisted surgery , 2004 .

[19]  F. W. Meredith.  The Modern Auto-Pilot , 1949 .

[20]  J. Bradshaw,et al.  Beyond Asimov : The Three Laws of Responsible Robotics , 2009 .

[21]  J. Horne,et al.  The impact of sleep deprivation on decision making: a review. , 2000, Journal of experimental psychology. Applied.

[22]  Alex Rosenblat,et al.  Algorithmic Labor and Information Asymmetries: A Case Study of Uber’s Drivers , 2016 .

[23]  Maarten Sierhuis,et al.  Coactive design , 2014, J. Hum. Robot Interact..

[24]  Niklas Strand,et al.  Semi-automated versus highly automated driving in critical situations caused by automation failures , 2014 .

[25]  J. W. Curtis,et al.  Passive switched system analysis of semi-autonomous systems , 2016, 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[26]  Raja Parasuraman,et al.  Adaptive Automation for Human Supervision of Multiple Uninhabited Vehicles: Effects on Change Detection, Situation Awareness, and Mental Workload , 2009 .

[27]  Vasile Dumbravă,et al.  Using Probability – Impact Matrix in Analysis and Risk Assessment Projects , 2013 .

[28]  S. Horgan,et al.  A prospective analysis of 211 robotic-assisted surgical procedures , 2003, Surgical Endoscopy And Other Interventional Techniques.

[29]  David E. Nye.  America's Assembly Line , 2013 .

[30]  A. Lanfranco,et al.  Robotic Surgery: A Current Perspective , 2004, Annals of surgery.

[31]  David Zuby,et al.  Collision Avoidance Features: Initial Results , 2013 .

[32]  Maarten Sierhuis,et al.  Human-agent-robot teamwork , 2012, IEEE Intell. Syst..

[33]  Andrea Maria Zanchettin,et al.  Safety in human-robot collaborative manufacturing environments: Metrics and control , 2016, IEEE Transactions on Automation Science and Engineering.

[34]  Markus Zimmermann,et al.  Vigilance, boredom proneness and detection time of a malfunction in partially automated driving , 2015, 2015 International Conference on Collaboration Technologies and Systems (CTS).

[35]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[36]  Michael Pilling.  Issues Regarding the Future Application of Autonomous Systems to Command and Control (C2) , 2015 .

[37]  James C. Christensen,et al.  Analysis of trust in autonomy for convoy operations , 2016, Defense + Security.

[38]  Huiyang Li,et al.  Human Performance Consequences of Stages and Levels of Automation , 2014, Hum. Factors.