Tesla’s Advanced Summon will be ready in ~6 weeks, says Elon Musk

(Credit: Tesla)

Tesla’s Summon feature might be pretty basic in its current iteration, but it already has a number of very compelling uses. One Tesla owner on Twitter, for one, used the function as a convenient and clever way to avoid parking penalties — something that Elon Musk found quite humorous. Discussing the feature in follow-up tweets, Musk provided a glimpse of some capabilities that are coming to the feature next year, as well as the estimated rollout date for “Advanced Summon.”

Musk noted by next year, Summon should allow vehicles to drive around a parking lot, locate an empty spot, recognize and read signs to confirm that a parking spot is valid, and initiate a parking sequence. For the upcoming months, though, Musk noted that “Advanced Summon” would be ready in around 6 weeks, and it would be rolled out through a simple over-the-air update. Advanced Summon would be compatible with all vehicles equipped with Autopilot Hardware V2+. 

Provided that Elon Musk’s timeline for Advanced Summon proves accurate, the updated feature would be rolled out to the fleet just before the 2018 holiday season. The timing of the feature’s expected release is quite notable, considering that Advanced Summon could very be a welcome end-of-year present for Tesla owners. 

With its upcoming improvements, Summon will become an even more invaluable feature of Tesla’s electric cars than it is today. Even at its current iteration, which only allows vehicles to move forward and backward for a maximum of 39 feet in a straight line, Summon is already a useful and novel function. A number of notable viral videos from the Tesla community involve the use of the feature. Veteran auto host Jeremy Clarkson found the function incredibly compelling during his review of the Tesla Model X P100D in the second season of The Grand Tour as well. 

Elon Musk has mentioned how future versions of Summon would work. In a tweet earlier this month, Musk stated that Summon’s upcoming iterations would use Autopilot cameras. This is a departure from the feature’s current system, which uses the vehicles’ ultrasonic sensors to maneuver in and out of parking spaces.

Elon Musk noted in his recent tweet that next year’s version of Summon should be able to read signs around parking lots to determine if a location is a valid parking spot. For this to happen, Tesla’s neural network, which continues to be refined and trained every day, should learn how to recognize and read signs. Current versions of Tesla’s software suggest that this is not the case yet. That said, during the third quarter earnings call, Tesla’s Director of Artificial Intelligence Andrej Karpathy noted that the company had actually trained large neural networks that work very well. They just can’t be deployed to the company’s current fleet due to limitations in hardware.

“It is a common finding in the industry, and that we see this as well, is that as you make networks bigger by adding more neurons, the accuracy of all their predictions increases with the added capacity. So in other words, we are currently at a place where we’ve trained large neural networks that work very well, but we are not able to deploy them to the fleet due to computational constraints,” Karpathy said.

Overall, Elon Musk’s Twitter update on Summon’s upcoming features seems to rely on the eventual rollout of Tesla’s Hardware 3 upgrade and the deployment of the company’s larger neural networks. Fortunately, Tesla Director of Silicon Engineering Peter Bannon noted in the past earnings call that Hardware 3 design is continuing to move along. Elaborating further, Bannon stated that validation for the upgraded hardware is already underway, and that the custom silicon, which is designed specifically for autonomous driving, is on track and “ready to go by the end of Q1.”

Simon Alvarez: Simon is a reporter with a passion for electric cars and clean energy. Fascinated by the world envisioned by Elon Musk, he hopes to make it to Mars (at least as a tourist) someday.
Related Post
Disqus Comments Loading...