Applicability of MIL-HDBK-516B to Certifying Autonomous Decision-Making Air Vehicle Systems

Alan Burkhard; Matthew Clark

doi:10.56094/jss.v51i2.159

Authors

Dr. Alan Burkhard Air Force Research Laboratory
Matthew A Clark Air Force Research Laboratory

DOI:

https://doi.org/10.56094/jss.v51i2.159

Keywords:

MIL-HDBK-516B, autonomous, aircraft, certification, software safety

Abstract

Airworthiness certification of military aircraft is accomplished by the developing military service. Air Force programs use the qualitative criteria outlined in MIL-HDBK-516B, “ASC/ EN Airworthiness Certification Criteria Expanded Version of MIL-HDBK-516B” (September 26, 2005) to aid the development of program-specific airworthiness criteria. The generalized criteria in this document are used to construct the specific criterion and associated artifacts — evidence of compliance — as the basis for making an airworthiness determination. This paper describes the process of transitioning from qualitative to specific criteria, and then examines the applicability of the existing guidance in MIL-HDBK-516B to autonomous decision-making adaptive air vehicle systems. Recommendations are made for future research and criteria expansion. An integrated approach that uses the most promising emerging and existing design, analysis, and validation and verification techniques is proposed as a means to develop the artifacts for certification coverage of autonomous adaptive unmanned air vehicle systems.

Author Biographies

Dr. Alan Burkhard, Air Force Research Laboratory

Dr. Alan Burkhard received his Ph.D. in engineering mechanics from the University of Wisconsin. He has more than 33 years of experience as an Air Force Research Laboratory researcher working on the development of testing methods and techniques for all types of mechanical, electrical and avionics subsystems. His experience includes being the chief engineer and team leader of a joint Navy and Air Force program for the development of integrated subsystems technology (JIST) for the Joint Strike Fighter. He currently is a Booz Allen Hamilton consultant to the Air Force Research Laboratory working on the development of verification and validation approaches for safety of flight software. He has authored and/or co-authored more than 20 technical papers and reports on test methods and criteria development.

Matthew A Clark, Air Force Research Laboratory

Matthew A Clark, MSEE is the technical area lead for the verification and validation of autonomous control systems within the Autonomous Controls Branch, Power and Control Division, Aerospace Systems Directorate, Air Force Research Laboratory (AFRL/RQQA). He leads a team of 10 in-house researchers in the design, analysis, verification and validation of autonomous control systems. He is also the primary subject matter expert for the AFRL Autonomy Test and Evaluation, Verification and Validation (TEVV). Additionally, he is the co-leader of the Assistant to the Secretary of Defense, Research and Evaluation, Autonomy Community of Interest, Test and Evaluation. He started his career in the Air Force Research Lab in 1995 supporting large-scale aircraft component thermal, acoustic and static combined environment structural testing. In 2000 and 2010, respectively, he received his bachelor’s and master’s Degree in electrical engineering at Wright State University with a concentration on electrical power and intelligent control systems. From 2000 to 2005, he worked as an industrial power and control engineer at Delphi Automotive in Warren, Ohio. In 2005, he returned to AFRL as technical area lead for the combined environment structural testing facility. In 2010, he served at the Air Force Material Command headquarters providing support for the test and evaluation infrastructure, strategic planning and operational cyber security, receiving the Exemplary Civilian Service Award. In 2011, he returned to the Air Force Research Laboratory to work on the verification and validation of autonomous control systems and applications. His research interests include verifiable intelligent power and control systems and run time assurance of intelligent systems.