Inside GNSS Media & Research

NOV-DEC 2017

Issue link:

Contents of this Issue


Page 41 of 67

42 Inside GNSS N O V E M B E R / D E C E M B E R 2 0 1 7 grams aim at achieving Level 3, although the mere presence of a kill-switch would imply that the system is actually Level 2. e transition from Level 2 to 3 is a remarkable leap that has significant implications on trust and comfort of human-machine interactions, on legal responsibility allocation between system and driver, and on technical challenges to overcome to guarantee passenger safety. Over the past four years, the most publicized approaches to demonstrate Level 2 HAV safety have been experi- mental testing campaigns by Google, Tesla and U ber. Google's approach to have HAVs drive millions of miles with minimal human intervention has been documented up until 2015. At this time, Google cars have autonomously travelled an impressive three million miles. Tesla's autopilot is reported to have driven more than 130 million miles – on highways only – before it caused a fatality in May 2016. In parallel, NHTSA reports about 3,000 billion miles travelled each year on U.S. highways by human drivers, with 30,000 deaths caused by traffic accidents; this corresponds to about one fatality in traffic accidents per 100 million miles driven in the U.S. But, this number accounts for incidents on all roads, in all weather conditions, and for all vehicle ages and types. us, a purely experimental, complete proof that HAVs match the level of safet y of human driving would take about 400 years at Google's current testing rate (of approxi- mately 250,000 test miles per year), and would still take many decades if the test- ing rate increased exponentially. is is assuming that no fatalities occur during that time, that no major HAV upgrade is performed, and that the testing environ- ment is representative of all U.S. roads. Thus, while an experimental proof is conclusive, it is not practical. Other, analytical, methods must be employed to ensure HAV safety. Research Challenges In HAV Navigation Safety Multiple technical aspects developed over decades for automated flying could serve as starting points for automated driving systems. Figure 2 shows research areas with overlap between aircra (in blue) and car (in yellow) applications. Figure 2 is not intended to give a com- prehensive list of all aspects of auto- mation, but instead, it shows example technical areas that can be addressed using similar methods in aviation and automotive applications (in the green area). For example: • performance standards set for so- ware, communication, and electron- ic equipment are already being com- pared for aircra versus cars in the NHTSA report by Q. D. Van Eikema Hommes, Additional Resources. • the design of aircraft cockpit has been continuously improved over the past few decades, especially for highly-automated Unmanned Air Systems (UAS) with a remote pilot "in-the-box"; few car manufactur- ers envision futuristic car interiors where humans do not participate in driving, but as long as human- machine interactions are needed, HAV SAFET Y SAE Level Name Description Human driver monitors the driving environment 0 No Automation The human driver performs all driving tasks at all times. 1 Driver Assistance Either steering or acceleration/deceleration task by the system; driver expected to perform all other aspects of driving. 2 Partial Automation Both steering and acceleration/deceleration tasks by the system; driver expected to perform all other aspects of driving. HAV monitors the driving environment 3 Conditional Automation The HAV performs all driving tasks under limited, pre- defined conditions, and can request the human driver to intervene and take over control. 4 High Automation The HAV performs all driving tasks under limited, predefined conditions, without the expectation of any human intervention. 5 Full Automation The HAV performs all driving tasks under all conditions. Table 1 Society of Automotive Engineer (SAE) International's Levels of Driving Automation FIGURE 2 Example Similarities and Differences in Future Automated Flying versus Driving. The figure shows technical aspects of future automation that may be common to aviation and automotive applications (in the center of the figure), versus others that are specific to each application (towards the edges).

Articles in this issue

Links on this page

view archives of Inside GNSS Media & Research - NOV-DEC 2017