Tesla vehicles have always had maps to assist human drivers. Recent emails reveal the extent to ... [+]
Newly-released emails between Tesla’s
For example, Tesla associate general counsel Eric Williams lists “unmapped roads” as a one of several “circumstances and events to [sic] which the system is not capable of recognizing or responding.” Reiterating content from the Owner’s Manual, Williams also refers to Full Self-Driving as being guided by “the mapped speed limit,” and the need for T-junctions to be properly identified in Tesla’s map data.
Williams highlights the relevant warning in the Owner’s Manual:
“CAUTION: Stop Light and Stop Sign Warning requires the vehicle’s map to know that a particular stop light or stop sign exists in a certain location. It may be unaware of new stop lights or stop signs that are not captured in the map data. Therefore, Stop Light and Stop Sign Warning may not detect all stop lights and stop signs due to potential map inaccuracies.”
Nonetheless, the Owner’s Manual also suggests that maps are not central to the operation of such functions:
“Although Traffic Light and Stop Sign Control primarily uses visual data received from the vehicle's cameras, greater accuracy is achieved when using the most recent map data.”
Pages later, the manual warns drivers that the lack of map data may cause the vehicle to “ignore” intersections:
“Even the most recent map data does not include all traffic lights and stop signs. Therefore, Traffic Light and Stop Sign Control relies heavily on the ability of the cameras to detect traffic lights, stop signs, road markings, etc. As a result, Model 3 may ignore an intersection that is blocked from the camera's view (for example, obstructed by a tree or a large vehicle or object, or located near a steep hill or sharp curve).”
Uncertainty about Tesla’s reliance on mapping data has existed for years. During Tesla’s 2019 Autonomy Day for financial analysts, CEO Elon Musk declared high-definition maps “a really bad idea.” But Tesla has long relied on mapping data short of “high-definition,” as evinced by its Owner’s Manuals and its communication with the California DMV.
Indeed, Musk recently tweeted that improvement of Tesla’s Smart Summon feature, based on GPS datapoints “will happen.”
The newly-published emails between Tesla and the California DMV, which shed further light on Tesla’s use of mapping, were obtained by PlainSite, a legal transparency non-profit.
In these emails, Tesla counsel Eric Williams stated definitively that the upcoming City Streets feature within the Full Self-Driving package would, “continue to be an SAE Level 2, advanced driver-assistance feature.”
This means both that the driver must be fully attentive to the driving task, and that Tesla is not testing self-driving cars on public roads in California.
Williams reassured the DMV that, “Tesla’s development of true autonomous features (SAE Levels 3+)...will not be released to the general public until we have fully validated them and received any required regulatory permits or approvals.”
The extent to which “true autonomous features” require maps, high-definition or otherwise, remains to be seen.
