Types of Robotics Papers | Hacker News Reader

Types of Robotics Papers

twitter.com

38 pointsreteltech9 comments

That's very good.

Note lack of "we built an agricultural fruit-picking arm that can survive five years of picking".

Remember Tesla's robotic charging arm? [1] That was a good idea. If Tesla ever gets unsupervised autonomous driving to work reliably in parking lots, that will be useful. Then cars can self-charge.

(The snake arm seems complicated, but it's not. The arm is just segments, cables, and discs. The motors are all in the base. So this can be built in a way such that damaged arm replacement is about as hard as gas pump hose replacement. Snake arm robots are not used much because they have a wear problem where the cables go through the discs. But that's for painting robots, which move constantly. A charging robot makes maybe ten moves a day.)

[1] https://youtu.be/uMM0lRfX6YI

Wireless charging is much better for the use case for autonomous charging than the robot chargers... it’s equally efficient with no moving parts, less prone to failure.

There's no way that wireless power transfer achieves the same efficiency as a wired connection.

The efficiency metric would need to take into account the number of cars it can charge at a given time and things like that also.

Qualcomm Halo is able to transmit about 3.5KW. They once got up to 20KW in a demo. The technology has been sold off to WiTricity. Tesla's biggest Supercharger model transfers around 250KW. Although Qualcomm has been announcing that technology since 2013, there seem to be no non-demo installations.

“A partnership between Cabonline, Jaguar, Momentum Dynamics, and Fortam Recharge are launching a wireless charging taxi fleet in Oslo, Norway. The fleet consists of 25 Jaguar I-Pace SUVs equipped with inductive charging pads rated at 50-75 kW. The pads use resonant inductive coupling operating at 85 Hz to improve wireless charging efficiency and range.” [1] https://spectrum.ieee.org/cars-that-think/transportation/adv...

"Are launching". "Will be". No follow up on how it worked, a year later.

Tesla later filed a different idea for patent that involved a charging connector at the bottom of the car and a movable plug in the ground of the charging station that could be moved in x/y direction to compensate for differently parked cars above it. From what I remember this system even included a connection for a cooling fluid that would have run through a heat exchanger in the car. That would allow to charge the battery even faster as the resulting heat could have been dissipated via a dedicated cooling system of the charging station.

https://electrek.co/2017/04/22/tesla-patent-automate-chargin...

That is a little odd. Teslas have a battery cooling loop. Electrek's diagram even shows it. That should handle kilowatts of heat easily. You'd figure pointing a big fan at the car, like on a dynometer, would work if you needed additional cooling. How much additional heat would ultrafast charging create, to justify heroic cooling solutions? Heat generated during charging is all waste heat, after all, reducing total charge efficiency. Not very green.

Reply on news.ycombinator.com