Mmwave sensor target data

Other Hardware Sensors
1

After a fair bit of reading on mmWave sensors I decided to order a DFR C4001 25m presence sensor to test. It works well and I am happy, however in my reading I saw that there are sensors that can track 3 targets simultaneously and give distance and angle and/or X and Y coords for the tracked target.

I am a little disappointed that the DFR unit does not seem to have this ability, although I am damn sure it makes internal use of these data, but I cannot seem to find a way to access it.

Which sensor would you recommend to do this, I can do the number crunching with X and Y coords or the polar equivalent of that with ease, so it is really just to find the right sensor for the job.

2

I assume "the ability" you reference is returning "distance, angle, x/y." How does DFR not have this ability? You know your location (0,0), you measure TOF (d = t * C), you know your angle and find (x, y) with (r = sqrt(aa + bb)).

Are you saying the boards you lament not having do this on-board?

3

You can use Texas Instruments' IWR6843ISK + MMWAVEICBOOST

4

There was quite a bit of discussion about the HLK LD2451 radar sensor, that gives distance and angles to several targets: Using HLK LD2451 radar for speed measurment

6

I lament the seeming inability to read this information from the board. The library for it, seems to offer the following feedback variables :-

getTargetEnergy()
getTargetNumber()
getTargetRange()
getTargetSpeed()

and I defy you to use those variables to define the location of the object.

That is not to say it is not possible, perhaps someone has knowledge of an alternative to

#include "DFRobot_C4001.h"

Regards
Donald Pittendrigh

7 
getTargetRange()

Use distance from getTargetRange() with its relative bearing/angle to find "location" which is relative to "you" [edit] Maybe getTargetEnergy() will show a center of mass?

Inability defied. What next?

I doubt DO "the others" give you any better object location in space and time?

8

There lies the rub, no getTargetAngle() the range is simply a distance from the sensor without which there is no resolution, bearing in mind, as I am sure you know, the sensor covers an angle of 100 degrees. If you know how to read the angle, then the game is on to be sure!!!

9

Getting an angle assumes a known bearing (of the source). You must set/create that known bearing (of the source) to get the relative angle (of the target).

That is where getTargetEnergy() might show you the center of mass. You could also sweep from "target lost" through "target acquired" to "target lost" and average the angle.

[edit] I read about some other mmwave modules with multiple antennae, so they could detect relative line of bearing by the arrival time of the reflected wave over the antenna array.

[edit]Hmm... there looks to be eight or two antenna on the DFR... or maybe that is just one big antenna. I don't know.

[edit] From this web site: https://www.robotics.org.za/SEN0610

FMCW (Frequency Modulated Continuous Wave) is a type of radar system based on frequency modulation of continuous wave signals. Unlike traditional pulse radar, FMCW radar continuously emits a series of continuous wave signals with gradually changing frequencies while simultaneously receiving the reflected signals. By analyzing the received signals, measurements of parameters such as distance, speed, and angle can be achieved.

I am feeling more confident the capability exists in three steps: detect + math + fun!

10

is the measure or the radar energy reflected back from the target, it encompasses neither centre of mass nor direction

11

What does it encompass, if not (a) mass (a size)?
Bats and the blind use echo location to navigate. Why do you deny it works?

(x0, y0) minus (x1, y1) gives direction (and velocity, if time is introduced).

12

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.