Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Hello, rear radar :-)

This site may earn commission on affiliate links.
Don't have much of an update. I did get a chance to test range of just the rear Bluetooth radio but still don't have the max range. Parked in the empty end of a large parking lot so there were no obstructions and walked as far as I could with the back of the car facing me before I ran out of space due to a fence. Distance was approximately 350 feet (just over 100 meters) and the rear bumper radio was still connected to the phone. The Tesla app also showed "Phone Key Connected" indicating good communication with the car. Checked the signal strength but it was bouncing all over the place so I don't have a good idea of what it was at that point.
 
  • Informative
Reactions: GoTslaGo
Mods - what's up with keeping my original thread title? Please change it to "Hello, rear bluetooth-antenna". Or "Goodbye, rear radar" :) LOL!

@Zarwin, I love your experimental nature. Please continue.

Anyone: Why would Tesla bother to install such a "chunk" in the rear bumper? Model X for instance, has tiny bluetooth antennas everywhere, also in the rear bumper, center. What's with the form factor? What are they building in there?

More to the point, why does it look like it was tacked on as an afterthought?
 
  • Like
Reactions: zmarty
Don't have much of an update. I did get a chance to test range of just the rear Bluetooth radio but still don't have the max range. Parked in the empty end of a large parking lot so there were no obstructions and walked as far as I could with the back of the car facing me before I ran out of space due to a fence. Distance was approximately 350 feet (just over 100 meters) and the rear bumper radio was still connected to the phone. The Tesla app also showed "Phone Key Connected" indicating good communication with the car. Checked the signal strength but it was bouncing all over the place so I don't have a good idea of what it was at that point.

Hrm... What if other sensors could be used to triangulate the location of things. For example, what if Bluetooth could be used to measure location and movement of other cars and even people with cellphones. Not accurately enough to determine the exact location but enough to verify that the object the vision system sees is infact a physical thing instead of say a shadow. I true mesh of all the sensors in the vehicle to make better decisions about the environment.

Someone also mentioned car to car communication of key data like the cars speed, heading and destination if navigating (does not have to be final destination for privacy reasons). The presence of emergency vehicles ahead could be relayed in every direction as an alert. Could also be used to communicate with the greater environment as the car passes through it. For example road conditions and construction could be broadcast. As well as businesses and open parking spots. I know this stuff doesn't exist today, but certainly someone is going to build it and standardize it for all vehicles. Would Bluetooth be a good tool for this? Cheaper then cellular, easily retrofit into cars and must have some form of Bluetooth now anyway. This would just be passive for older cars but could be updated to be active by broadcasting telemetry to the surrounding area. Just a crazy thought.
 
  • Helpful
Reactions: croman
Hrm... What if other sensors could be used to triangulate the location of things. For example, what if Bluetooth could be used to measure location and movement of other cars and even people with cellphones. Not accurately enough to determine the exact location but enough to verify that the object the vision system sees is infact a physical thing instead of say a shadow. I true mesh of all the sensors in the vehicle to make better decisions about the environment.

Someone also mentioned car to car communication of key data like the cars speed, heading and destination if navigating (does not have to be final destination for privacy reasons). The presence of emergency vehicles ahead could be relayed in every direction as an alert. Could also be used to communicate with the greater environment as the car passes through it. For example road conditions and construction could be broadcast. As well as businesses and open parking spots. I know this stuff doesn't exist today, but certainly someone is going to build it and standardize it for all vehicles. Would Bluetooth be a good tool for this? Cheaper then cellular, easily retrofit into cars and must have some form of Bluetooth now anyway. This would just be passive for older cars but could be updated to be active by broadcasting telemetry to the surrounding area. Just a crazy thought.

I think the infrastructure side could be something Starlink might support. Classical Bluetooth is not so great at dynamic pairing due to the discovery method, number of active connections, and provider/consumer topology. Beacons might be able to provide the data you speak of.
I'd personally worry about a system that relies on BT for object detection since that could imply it won't avoid you if you phone battery is dead. (but I'm also firmly in the camp of, "If it doesn't work in the worst case, it doesn't work")
 
I think the infrastructure side could be something Starlink might support. Classical Bluetooth is not so great at dynamic pairing due to the discovery method, number of active connections, and provider/consumer topology. Beacons might be able to provide the data you speak of.
I'd personally worry about a system that relies on BT for object detection since that could imply it won't avoid you if you phone battery is dead. (but I'm also firmly in the camp of, "If it doesn't work in the worst case, it doesn't work")

I would never say rely on, only add to. There is an issue of having to much data as well. Its expensive to process and there could be a lot of contradictions that cause deductions to be worse then they would have been without so much information. It would only make sense if it was something that would reliably add to the total sensor mesh. You dont necessarily have to handshake for it to contribute. Just detecting a BT single and being able triangulate and track its motion enough to determine direction and speed would be useful. Is there no such thing as a broadcast for BT that does not require a handshake to share information?
 
  • Like
Reactions: mongo
Is there no such thing as a broadcast for BT that does not require a handshake to share information?
The beacon mode is (usually) a one way, no handshake needed protocol. Synchronized receivers can triangulate position from a discontinuous source via differential time of flight. (what time does the pulse reach each different receiver).

Understood regarding using all data available :) Lots of time/ effort/ complexity/ reliability trade offs.
 
Holy cow this thread.

Ok, so, it is almost certain that the BLE (Bluetooth Low Energy) modules on the side pillars and rear bumper are used for the BLE PXP (proximity) profile. This is a proximity sensing profile that allows devices to detect the presence of another BLE device. This also gives me hope for a feature I'd love to see- unlock the mobile device to unlock the car. Because the sensing of the presence of a device allows the mobile OS to present notifications or other actions, it could easily request that iOS or Android force the user to unlock the device to allow access to the car. Tesla: A simple checkbox to enable/disable this would be great. Thanks.

As for the number of sensors, this could be for lots of reasons. It could be useful for RSSI (Received Signal Strength Indicator) filtering which is roughly how proximity is calculated, it could be useful to detect where a user is approaching from, it could be useful to verify a sensor isn't being attacked with a directional antenna of some kind, it could be useful for detecting proximity in an area with lots of devices or signal noise, I could think of a thousand reasons. And, since the sensors are so cheap, ($2.15 or less per thousand, with progressive discounts as volume increases), there's basically no reason not to add a bunch of them. For a million cars, and the maximum price, these are still only $2,000 total component cost.

For the people thinking Tesla would use BLE for radar. No. Just no. Stop it. The wavelength for 2.4GHz is 12.5cm. Bosch (and Continental) MRR/LRR systems use 76GHz frequency which is a wavelength of 0.39cm or 3.947mm. The resolution is almost 32x better. Not only that, but the noise on the 2.4GHz band in a car that uses bluetooth, with a phone that scans for WiFi, on roads with cars that have WiFi hotspots in them driving by, it would be a signal filtering nightmare. And the chips involved simply do not have that kind of processing power or features onboard. The datasheets are freely available from Texas Instruments here. Check out the device and the development kits available. They're super simple devices that do a good job at one thing. Bluetooth LE.

One big positive here, this does mean at some future date Tesla could potentially build a FOB or a very small pocket bluetooth device that could replace a phone or the card key. The batteries in Bluetooth Smart or low powered BLE devices last for years at a time. It's basically how Tile and similar products work. If you want to learn more about PXP profile or other BLE GATT (Generic ATTribute) profiles, see the bluetooth.org GATT page here. There are tons of them, and lots of them have very cool features that can be implemented at the same time. This is all pretty "boring" standard stuff that has existed for a long time. Tesla, Apple, and Google/Android are simply adopting these existing technologies better and faster than everybody else, and using them in interesting ways. This is much better than a rear-facing radar system, because this vastly improves the user experience of the vehicle!

Edit: I defined abbreviated terms, because not everybody here is a software or hardware developer. Sorry.
 
Last edited:
If you want to learn more about PXP profile or other BLE GATT (Generic ATTribute) profiles, see the bluetooth.org GATT page here.
Funny you mentioned that, my previous hobby project was reverse engineering a BLE-controlled cooktop appliance (that only had iOS app) and writing an Android app for it.

For those interested to perform it on a Tesla, get the BlueFruit (or from Amazon) and play with it.

Also you can probe and experiment with attributes just from your phone, get nRF scanner or LightBlue Explorer and connect it to your car device(s) and see what's exposed and what happens when you write to writeable attributes and what changes in read-only attributes as you do things with the car.
 
My guess. These sensors are going to be used for the Tesla Network somehow. Maybe to communicate with Superchargers so the car can charge itself without driver/passenger intervention. Honestly I think that may be the difference between EAP hardware 2.0 and 2.5. 2.5 cars will be able to "rent" their cars on the Tesla network and hardware 2.0 won't be able to.....obviously just my crazy speculation.
 
My guess. These sensors are going to be used for the Tesla Network somehow. Maybe to communicate with Superchargers so the car can charge itself without driver/passenger intervention. Honestly I think that may be the difference between EAP hardware 2.0 and 2.5. 2.5 cars will be able to "rent" their cars on the Tesla network and hardware 2.0 won't be able to.....obviously just my crazy speculation.

Tesla would be in trouble because it said 2.0 cars would be able to do that. Not the first time. Anyways, I don't see why they'd need this. The car's identity is already known to Tesla when supercharging. There is a handshake that occurs and authorization to charge.
 
Tesla would be in trouble because it said 2.0 cars would be able to do that. Not the first time. Anyways, I don't see why they'd need this. The car's identity is already known to Tesla when supercharging. There is a handshake that occurs and authorization to charge.
True but how do you get the charger to line up with the port? Maybe some wireless transmission to let it know its there before plugin?
 
Don't have much of an update. I did get a chance to test range of just the rear Bluetooth radio but still don't have the max range. Parked in the empty end of a large parking lot so there were no obstructions and walked as far as I could with the back of the car facing me before I ran out of space due to a fence. Distance was approximately 350 feet (just over 100 meters) and the rear bumper radio was still connected to the phone. The Tesla app also showed "Phone Key Connected" indicating good communication with the car. Checked the signal strength but it was bouncing all over the place so I don't have a good idea of what it was at that point.
Large BlueTooth antenna for maximum range when using the summon feature from across a parking lot?
 
this rear radar doesnt help with forward cross traffic which is a bigger problem

Why is forward cross traffic a problem at all? Isn’t that why we have stop signs and traffic lights?

At an intersection with only two stop signs, forward cross traffic would be a problem. But the car has 360-degree cameras for that. Is the forward radar beam too narrow to catch those cars also?
 
Why is forward cross traffic a problem at all? Isn’t that why we have stop signs and traffic lights?

At an intersection with only two stop signs, forward cross traffic would be a problem. But the car has 360-degree cameras for that. Is the forward radar beam too narrow to catch those cars also?
You want to see in case others are going to drive into you. What if a truck starts pulling across a highway? Remember, if a machine doesn’t avoid a human making a mistake, the machine may get the blame. Need to be better.
 
True but how do you get the charger to line up with the port? Maybe some wireless transmission to let it know its there before plugin?
Or we could use the most "advanced AI" and neural networks with a camera.... Which is the same thing powering the car..... And have I told you how it performs almost 10x better than a human in my secret test lab?
 
  • Like
Reactions: DrDabbles
True but how do you get the charger to line up with the port? Maybe some wireless transmission to let it know its there before plugin?

Bluetooth wouldn't really be able to be used to line something up with a charger port. It's a data network. But, BLE could be used for advanced features that couldn't be offered from the data transmitted over the charger bus.

Large BlueTooth antenna for maximum range when using the summon feature from across a parking lot?

...which would also mean your car would unlock when you're across a parking lot. BLE already has an outdoor range of 100m, and relies on the RSSI to determine how far away your phone is. Also, there's the small legal matter of the FCC limiting EIRP from wireless devices in the unlicensed ISM bands that Bluetooth and WiFi operate in. So a "large" antenna would basically be illegal and cause major interference issues with the car receiving far away signals, and the car blasting other devices with overly powerful signals.

Triangulation?! Could their beams be used to position the car itself more accurately in an HD map than vision can do, or to harvest/collect any geometry. I am not expert.

It's Bluetooth. No.