Better safety through radar

Zabhawkin

Well-known member
Joined
Sep 1, 2021
Threads
11
Messages
323
Reaction score
529
Location
New Mexico
Vehicles
1999 Nissan Frontier, 2015 F-150, 1984 Jeep CJ7
Country flag
The problem I have with active systems is there is only so much band width to work with, either by legislation, or what frequency ranges can actually accomplish the job. The more and more cars that get on the road using active systems the more noise they have to deal with and the systems become degraded just by proximity.
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Threads
126
Messages
16,229
Reaction score
27,097
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
The topic is important, but they made a basic error with their first example:

30% chance of rain means 30% of the forecast area may see rain or 30% of the forecast time will see rain. In Seattle, 30% chance is more likely to mean the latter while in Phoenix it's more likely to be the former.

Where I live, if there is a system with 30% bands of rain, that's nearly 100% chance of rain - because our hill will wring the water from the clouds, it's the 30% of area. But if the report is 30% of scattered thunderstorms (like you get in Phoenix) then well, there's less than a one in three chance we'll even see the storm clouds, let alone hit the hill.

It also doesn't take into account the intensity of rain - Seattle is known for light showers while you literally wouldn't go outside in a thunderstorm in Phoenix, let alone with an umbrella.

And the people's answers already calculate that into the risk. 😏

-Crissa
 
Last edited:

Zabhawkin

Well-known member
Joined
Sep 1, 2021
Threads
11
Messages
323
Reaction score
529
Location
New Mexico
Vehicles
1999 Nissan Frontier, 2015 F-150, 1984 Jeep CJ7
Country flag
It also doesn't take into account the rain - Seattle is known for light showers while you literally wouldn't go outside in a thunderstorm in Phoenix, let alone with an umbrella.


-Crissa
What you don't want to do a Mary Poppins impression? (known for heavy wind gusts near thunderstorms)
 


CyberGus

Well-known member
First Name
Gus
Joined
May 22, 2021
Threads
67
Messages
5,817
Reaction score
19,101
Location
Austin, TX
Website
www.timeanddate.com
Vehicles
1981 DeLorean, 2024 Cybertruck
Occupation
IT Specialist
Country flag
30% chance of rain means 30% of the forecast area may see rain or 30% of the forecast time will see rain. In Seattle, 30% chance is more likely to mean the latter while in Phoenix it's more likely to be the former.
In Texas, a 50% chance of rain actually means a 100% chance of rain on every-other house.
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Threads
4
Messages
3,213
Reaction score
3,403
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
30% chance of rain means 30% of the forecast area may see rain or 30% of the forecast time will see rain. In Seattle, 30% chance is more likely to mean the latter while in Phoenix it's more likely to be the former.

Where I live, if there is a system with 30% bands of rain, that's nearly 100% chance of rain
In Texas, a 50% chance of rain actually means a 100% chance of rain on every-other house.
Insight comes from the strangest places. These posts make it clear than the frequentist approach to probability just doesn't work for things like weather forecasts. And indeed NWS's definition uses Bayesian language. They say a 30% probability of rain means a 30% chance of 0.01" or more within a 12 hr period in a 5 km square.
 
Last edited:

ituner-HF

Well-known member
First Name
Andrei
Joined
Feb 1, 2021
Threads
39
Messages
118
Reaction score
232
Location
Pleasanton, CA
Vehicles
S, X and 3, Waiting for Cybertruck
Country flag
Lidar?

Radar may or may not work okay in rain.

The problem with radar is that it's weird, compared to sight. It bounces around, it goes through lots of things like people, etc. You're trying to detect the world through something that finds much of the world varying levels of transparent and refractive.

-Crissa
The wavelength of LIDAR is susceptible to rain and snowflakes, RADAR operates at a much lower frequency (70-80Ghz), wavelenght is simply too long for rain detection, it goes through it with min attenuation.
 


ituner-HF

Well-known member
First Name
Andrei
Joined
Feb 1, 2021
Threads
39
Messages
118
Reaction score
232
Location
Pleasanton, CA
Vehicles
S, X and 3, Waiting for Cybertruck
Country flag
Safety is what I want most from my Cyber Truck. (I want all the other great features too.) Therefore I want it to see what I can’t see with my eyes that could keep me safe when human vision can’t detect what is happening. The Cyber Truck should see in/through the fog, through the spray of water that comes up from a semi or other vehicles on the rainy crapy roads and freeways of the pacific northwest. I want it to see through the snow flying up from behind another vehicle so common in winter in cold climates like Montana etc. In order to do this it needs some kind of active vision like radar or lidar. If the FSD only works in fair weather I can do that, I want my amazing machine to do what I can’t do. To rely only on regular cameras is limiting. Unless Tesla can tell me otherwise that their camera can see through fog etc I want it to have radar or something to see what human vision can’t.
I just wanted to say that I am impressed by latest 4D radar with MIMO antenna array, it is quite impressive. Tesla's next gen radar unit (as Approved by FCC last week) seems to be 4D radar. It is almost as good as LIDAR, sans interference from rain or snow. Such a small unit!

This is an exmaple of a 4D radar unit with MIMO.


Tesla Cybertruck Better safety through radar 1655960386306
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Threads
126
Messages
16,229
Reaction score
27,097
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
The wavelength of LIDAR is susceptible to rain and snowflakes, RADAR operates at a much lower frequency (70-80Ghz), wavelenght is simply too long for rain detection, it goes through it with min attenuation.
That's why I said 'may or may not'. Radar that's good at picking out small objects and people is more susceptible to rain interference.

-Crissa
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Threads
4
Messages
3,213
Reaction score
3,403
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
RADAR operates at a much lower frequency (70-80Ghz), wavelenght is simply too long for rain detection, it goes through it with min attenuation.
Don't know if this is a typo but the earliest radars were in the high HF and low VHF bands i.e. that 70 should be 70 mHz. The thing that turbed Watson-Watt's idea into a reality was the availability of TV tubes (or in his case "valves"), I suppose the first real radar was the British experimental sets that worked at 6 MHz.

If the illuminating wave length is much longer than the diameter of a raindrop it will not be scattered but at, for example, X-band rain clutter is indeed a problem. As mentioned in an earlier post shifting to the use of circular polarization or the use of a lower frequency helps amd doppler filtering does too.
 

rr6013

Well-known member
First Name
Rex
Joined
Apr 22, 2020
Threads
54
Messages
1,680
Reaction score
1,620
Location
Coronado Bay Panama
Website
shorttakes.substack.com
Vehicles
1997 Tahoe 2 door 4x4
Occupation
Retired software developer and heavy commercial design builder
Country flag
I just wanted to say that I am impressed by latest 4D radar with MIMO antenna array, it is quite impressive. Tesla's next gen radar unit (as Approved by FCC last week) seems to be 4D radar. It is almost as good as LIDAR, sans interference from rain or snow. Such a small unit!

This is an exmaple of a 4D radar unit with MIMO.


1655960386306.png
@ituner-HF what about processing? Speed of PUREvision ended up waiting on RADAR artifacts stuck in process which pushed radar off the car in the first place? That gave back performance at the expense of FSD learning how to stop phantom braking. What? SO FSD can’t cure phantoms in realtime?

Perhaps… Tesla is now redeploying RADAR controlling in a parallel architecture with PUREvision. Phantoms would pre-flag on the RADAR pipeline against which PUREvision would verify its each and every phantom occurrence on its stack. Matches on both activate EBS else NOP, continue as if nothing was there, ‘cuz RADAR.

Anyone? have a clue such a scenario even possible in realtime? Otherwise Tesla FSD is back to latency delayed processing burdens.

A reverse scheme PUREvision controlling; computer scientists could have developed a parsing filter for radar, mapping it against PUREvision’s FoV selecting for only elements “Not A Match“ to join PUREvision main process stream. That would eliminate all elements both systems “see” still leave, remaining, a much smaller workload to compare only two NIL spaces on-chip.

Something is up! Andrew Karpathy took himself on vacation from “Project Vacation” - DOJO. RADAR is back(nod to Schwartzneggar)! And NTSB is sniffing.
Sponsored

 
 




Top