Inside Unmanned Systems

DEC 2018 - JAN 2019

Inside Unmanned Systems provides actionable business intelligence to decision-makers and influencers operating within the global UAS community. Features include analysis of key technologies, policy/regulatory developments and new product design.

Issue link: https://insideunmanned.epubxp.com/i/1062779

Contents of this Issue

Navigation

Page 38 of 59

39 December 2018/January 2019  unmanned systems inside and we should consider this when designing anten- nas in this industry." SAFETY OF AUTONOMOUS SYSTEMS Khanafseh finished out the webinar with a discus- sion on safety. The motivation behind autonomous driving is that it's safer than manned driving. It's expected to lead to a lower fatality rate, but that doesn't, however, imply autonomous systems should be allowed to be tested, and to count the fatality rate. "Instead there should be enough reliability al- ready built into the safety of the autonomous driving system," he said. "Certain design requirements and safety standards with documentation of the system proof of safety should be made in front of a certifica- tion entity. You can think of it as just like a driver's license. You just don't self-proclaim yourself to be a driver and then just start driving on the streets." It's not enough to demonstrate a system's safe- ty case through testing and demos only, he said. While they're helpful, many questions surround them such as testing environment, were there enough tests and was the likelihood of "hazardous" or "catastrophic" events quantified beforehand. Khanafseh suggests borrowing concepts from the aviation industry to provide design safety. The questions we're asking now were common in that industry years ago with different navigation systems, especially GNSS. Aviation has defined quantifiable system performance measures that include accuracy, system integrity, continuity and availability. While he didn't have time to go into detail about each performance measure, the take- away is the idea that an entity puts requirements on these systems, and designers or builders must provide a case proving whatever system or design they come up with meets those requirements. He also talked about integrity risk, which is a measure of how much the position provided by the navigation system can be trusted. This is ap- plication dependent, but we must ask if a wrong position estimate will cause any financial or life risk, or if an interruption to the operation/false- alarm will cause any financial or life risk. If the answer is yes, we must provide a level of accept- able likelihood for a wrong position/interruption. Threats that are sufficiently likely to occur will require a monitor to mitigate the threat effects. Khanafseh went on to discuss integrity risk evaluation, key aspects of navigation system in- tegrity, GPS faults/vulnerabilities, fault mitiga- tion and the integrity of GPS versus other systems. The webinar ended with a second question and answer session. To hear the answers to par- ticipants' questions and to learn more from the expert panelists, you can download the webi- nar at https://attendee.gotowebinar.com/regis- ter/1238704696436627971. BY THE NUMBERS » PARTICIPANTS´ VIE S What level of positioning accuracy do you think autonomous driving requires? In your opinion, what is the biggest risk to integrity? How critical do you think the GNSS component is compared to other autonomous driving sensors (I.e. LiDAR, radar, camera, etc.)? Public safety 65% Cost 22% Regulation 43% Infrastructure 27% What do you think the biggest barrier is to widespread production of autonomous vehicles? (select your top two) Technology limitations 30% Somewhat important 12% Not important at all 1% Very important 65% Most important 22% 0–10 cm > 1 Meter 1% 50 cm– 1 Meter 9% Jamming and spoofing Data security 4% GNSS measurement errors Interference 44% 10–50 cm 46% 28% 30% 38%

Articles in this issue

Links on this page

Archives of this issue

view archives of Inside Unmanned Systems - DEC 2018 - JAN 2019