I have a vehicle with a tracker installed. The device has a gps syst, 3-axis accelerometer, 3-axis magnetomet and a gyromeyet. Is it possible to determine by how much the vehicle rolled back on a slope or hill. Using gps angle wasn't an option as the angle given for short backward movement isn't always reliable. Can accelerometer be used in such a scenario??
You're right that the GPS angle (heading) will not help you in a single-antennae setup. On its own a GPS receiver needs a minimum distance of movement to determine heading.
A simple GPS receiver, when used without GPS corrections (which is the case for off-the-shelf GPS devices and mobile phones/tablets), has a minimum ~5 meter accuracy. That's why a short backward movement will not yield the desired results.
In construction/mining applications, there is often a fixed GPS base station nearby that broadcasts GPS corrections, which allows vehicle-mounted GPS receiver to apply corrections, reduce error and ultimately get centimeter-level accuracy.
So in conclusion, your 3-axis accelerometer will likely be the only sensor that you can rely on until your vehicle has rolled back at least 5 meters.
If your accelerometer is sensitive enough, you'll get measurable sensor values. However, if you rollback is very slow, where the G forces are almost imperceptible to the accelerometer, then you're out of luck.
This is assuming that you want near real-time detection of vehicle rollback.
Related
I am wondering if it makes sense to fuse multiple GPS signals to improve my estimated result. This works fine for example for accelartion sensors, but this sensors have a white gaussian noise.
GPS sensors being mounted on the same board probably suffer from the same errors like drift or multi-path effects, which cannot be corrected by only fuse the sensor readings of this sensors. I imagine that like a constant offset in the same direction, which won t be correct just stays nearly the same.
Furthermore, I have diffrent sensor which I can mount on my drone, even RKT sensor. In my opinion, it makes no sense to fuse a d-GPS with readings from an RKT GPS.
Please correct my if I am wrong.
Thank you in advance and I hope this forum is the right spot to ask that question.
yes you can. Use EKF based approach with onboard multi GPS and multi IMU
The DJi is doing it, But it is can only prevent one of sensor failure, not the systematic drift patter. To avoid that, you need some more source such as visual odometry or lidar odometry to fuse in the EKF. GPS sate count is good meaure of how bad the position is. It ranges from 0 to 15. So when every one is 15, trust GPS more less variance. When everyone is lower than 6 add very high variance to GPS source.
Yes RTK might be better when you have direct line of sight. But once out of sight, then other GPS might be better. So totaly depends on your use case
There are a fair number of GPS questions here...
The GPS is not always completely accurate.
If two devices are nearby, will they have the same error?
I have GPS in my lawnmower. If an app subtracts the GPS location of the mower and the GPS on the phone, will the error cancel out, so it can show the direction to the mower with greater accuracy than just looking at the GPS location of the mower on the map?
No, the error can be different even if the devices are of the same model.
The error is due to signal diversity which can change even if you move the receiver even a bit, location of the antenna (which is definitely not the same) and the accuracy of the GPS's components. All of these can vary even for two devices of the same model. If the models are not the same, they can be using different technologies, like SiRF3 or SiRF4 so once again, the error probably not be the same.
I'm using a high accuracy GPS RTK setup to precisely locate a mobile robotic platform in the field (down to 10 cm accuracy). I have also a 9DOF IMU mounted on the platform (9DOF sparkfun IMU Razor).
The Question is, Do I really need to perform a sensor fusion between IMU and GPS like what this ROS node do (http://wiki.ros.org/robot_localization) to estimate the robot pose? or is it just enough to read the Pitch,Yaw,Rotation data from the IMU to know the heading along with the GPS Long,Lat,Alt ?
What cases make it essential to perform this type of fusion ?
Thanks in advance
It is essential to perform fusion because:
1) Roll, Pitch, and Rotation data from the IMU are not perfect, and they will drift over time due to gyro errors. The magnetic field sensor in the IMU module limits this, but crudely. Fusion allows the GPS RTK measurements to be used to continuously estimate the dominant error sources in the IMU and maintain better attitude information.
2) The IMU supports position estimation when GPS-RTK is lost through signal blockage or any other outage, such that the robotic platform is not lost when and if GPS signals are interrupted.
I just bought Digital heart beat rate sensor:
http://www.dealextreme.com/p/digital-heart-beat-rate-sensor-3-5mm-data-port-16009
And I'm looking how I can make application for iOS to work with.
Sensor has 3.5mm jack and I can detect signal with audio framework on iOS.
Can you give me some guidelines how to start with detecting these signals into heart beat rates?
That sensor looks rather like one I have here in my junk box. If so, it generates a voltage signal which depends on the pressure exerted on it by the skin against which it is pressed. If there is a strong pulse at the point of pressure, I see a signal on an oscilloscope which has a component at the pulse rate: so it is at a frequency of around 1-2Hz.
This is WAY below the audio range, and in most audio interfaces would be filtered out before it ever got to the audio in ADC. I don't have a handy iPhone to check this on, but it would be bad design if the audio input did let such frequencies through. And Mr Jobs (R.I.P.) did not approve of bad design!
There is also a lot of interference at other frequencies: mains hum (50Hz here), and at lower frequencies spurious signals from muscle twitches.
To make this work, you would need some sort of signal conditioning. If it was up to me, I would use a high input impedance amplifier, with about a 0.1Hz - 10Hz passband, followed by a voltage to frequency converter. That would give me a tone, which i could set in the audio band, whose frequency varied up & down as the pressure on the sensor changes. That would let me use fairly simple frequency detection software to recover the pressure waveform, which could then be processed using autocorrelation or similar techniques to recover the heartbeat frequency. A DTMF decoder is not the right tool, though.
I did find when I played about with the senor that it was very touchy, responding to almost everything going, and it wouldn't be easy to pick out the heartbeat. Your sensor may be different, though.
i want to detect heart rate using iphone sdk does someone knows any method for calculating heartbeat rate?
Fast Fourier Transform is a class of algorithms that can quickly turn samples into an analysis that tells you how prominently ceratin frequencies occur in that sample. For more check out:
Wikipedia: FFT
Literate program example: Cooley-Tukey FFT
This is relevant to your problem because: (1) heart rate is itself a frequency, and (2) most of the sound that comes through the body that you can measure will be within a certain frequency range. Dropping frequencies outside this range means dropping all or mostly noise.
Good luck!
Well I've seen various implementations. Some of them use the accelerometer to detect minute movements in your arm/hand when you hold the phone, some of them can use the microphone, you could also do a manual 'tap' interface where you tap the screen while checking your own pulse.