Three lessons the FAA might gain by Tesla’s automobile accident
In most NTSB investigation proceedings, the FAA usually is a party to the effort and is often the subject of one or more of its findings. The current Board examination of the May 7 crash involving a Tesla Model S driving on Autopilot may be a forum in which the FAA might consider observing carefully.
Functionally and analytically the Tesla “autopilot” is not the same as the equipment found in modern aircraft cockpits. The systems which drive those planes are far more sophisticated than the automotive equivalent. The aircraft’s equipment deals with three dimensions, far greater speeds, an almost infinite number of approaches and connections with real time data generated by radars. In fact, one commenter argues that the Tesla instrument should be more accurately described as “driver assist mode.” So, why would the FAA be interested in this automobile case; here are three useful lessons which might be gained by this accident review:
- One of the likely lines of inquiry by the NTSB might be the driver-machine interface. The FAA and the Board are both concerned about the pilot-machine connection and the consequences of removal of the human from that guidance system. Experts will be called in to testify on the values and risks associated with the May 7 crash. Their insights may be relevant to the FAA’s safety mission and associated knowledge.
- The trade press is heralding the advance of automobiles which can fly. Soon, the FAA will have to certificate these cars for their flight mode. Those land/air vehicles will likely have some system to assist the land driver in the plane mode. The connectivity between the ground and aerial systems will be of interest to the FAA and the basic Tesla system’s specifications and functionalities might be the electronic foundation for the aerial application.
- The FAA’s most pressing technological issue is the UAS “sense and avoid.” The existing autonomous control of commercial and GA aircraft is large, complex and expensive; so it is likely that the sense and avoid systems for drones will not “dumb down” the larger airplane systems, but may well build off of the more compact, less costly Tesla equipment or other automotive “driver assist mode” equipment. Further the car system does not rely on radar input or controller communications; so the search algorithms may be more comparable for drone until there is a UAS ATM. The testimony and evidence produces at the NTSB’s hearing may provide very relevant information.
Curiously, the FAA may be able to treat this NTSB investgation as a useful tutorial even though the focus is not planes; investigating automotive guidance systems may provide technical information transferrable to three FAA high interest areas.