Tesla’s New Autopilot needs more improvement : Consumer Reports

10 Oct 2016

Though Tesla Motors has incorporated several new safety features in its Autopilot, including tighter limits on how long drivers could keep their hands off the wheel, its semi-autonomous driving system still raised real concerns, says non-profit Consumer Reports, published by consumer advocacy organisation Consumers Union. Consumer Union is also involved in product testing, consumer-oriented research and public education.

The findings relate to the Tesla Model S test car in Autopilot mode after the car received a software update called Tesla 8.0. (Tesla, like Apple also updates its cars wirelessly). Besides several changes to Autopilot, the update also comes with additional warnings for drivers to keep their hands on the wheel. In case repeated warnings are ignored by the driver, the car is forced to restart.

In July, Consumer Reports had asked Tesla to make a number of changes to Autopilot, after an evaluation of an earlier version of the semi-autonomous system.

Meantime, Tesla's Autopilot had come under the scanner of regulators following a fatal accident in May. The accident is being investigated by the National Highway Traffic Safety Administration and National Transportation Safety Board.

Following the fatal accident which killed the driver, there have been debates whether the company's ''self-driving'' cars were safe (See: Florida death puts big question-mark on Tesla autopilot ).

On 4 October the company was asked by California's regulators to stop advertising its cars using the word ''autopilot'' until and unless the car could actually drive itself, without human intervention.

Currently drivers are required to stay awake and attentive even with the autopilot system on.

Meanwhile, even as the controversy about the system safety continues to swell an arfticle in The Verge said , ''Tesla says that Autopilot has driven 130 million miles in owners' vehicles, now with one fatality; that compares to a US average of one vehicular fatality every 94 million miles. So yes, it is statistically doing better than average, but there's an expectation - however fair or unfair - that the computers can and should be perfect.''

But, according to commentators whether Tesla's Autopilot was ''better than average'' and had prevented more accidents than it caused were at the core of the issue.

They say, features like adaptive cruise control and lane departure warning would certainly make driving much safer at some point in the future, but whether they could do so immediately was a question that was far more complicated.

They point out that while ''safe'' conclusion might or might not be sound, the argument the system's supporters used to back it up clearly was not statistically significant.

They point to a RAND Corporation paper issued in April which said, ''Autonomous vehicles would have to be driven hundreds of millions of miles and sometimes hundreds of billions of miles to demonstrate their reliability in terms of fatalities and injuries.''