Hi, Jimit here, I recently build Extended Kalman Filter for Attitude estimation, but I don't know how to perfectly tune it! what are the Q matrix and R matrix, and which value I can put in diagonally, are they found in Sensor's datasheet? If yes then Which value in the datasheet do I need to put from the datasheet? And if not then what are the values, and how to find optimal values? I perfectly know what are Q and R matrices, we add noise in our prediction using the Q matrix, and similarly, we add measurement noise using the R matrix, and Kalman gain decides when we trust our prediction or measurement, I know all about it maths, how the filter works, but I need to know how we find R and Q matrix's value in the real world, how engineers find them!
One place to start is to measure the sensor noise.
I am doing Alen deviation, and finding angle random walk. what should I do now, can I put this variable into my Q and R matrix? please guide me brother!
Perhaps you understand the theory but lack practical understanding of the details needed to implement it. If that is the case, an EKF is certainly the wrong place to start.
Start with a simple 1D linear Kalman filter for the sensor, and learn how to scale the measured (or the data sheet values of) sensor noise to the data, in order to get the expected result out of the filter measurements.
Kalman filter theory includes sources of noise that might not be directly measurable, and you need to be able to estimate reasonable values to use in the implementation. Trial and error works in many cases.
Hi bro, I also implement ekf in my previous project where I tune my ekf, I set initials values for Q and R matrices, then I start tuning, what I am doing here is I run drone motor I tune until it has no effect for vibration, when I get point where vibration less influence results and angle not drift in that case I stop tuning and save my value, In practical it is not good idea I know, so what is right tricks, please inform me, I need to know for my drone project, I use quaternion for attitude estimation, I also estimate gyro bias I am so curious for this, I have passion to learn it, please guide me
Your other topic on the same subject deleted.
Please do not duplicate your questions as doing so wastes the time and effort of the volunteers trying to help you as they are then answering the same thing in different places.
Please create one topic only for your question and choose the forum category carefully. If you have multiple questions about the same project then please ask your questions in the one topic as the answers to one question provide useful context for the others, and also you won’t have to keep explaining your project repeatedly.
Repeated duplicate posting could result in a temporary or permanent ban from the forum.
Could you take a few moments to Learn How To Use The Forum
It will help you get the best out of the forum in the future.
Thank you.
can I be doing that? I record flight data record all sensor data, then I do EKF in Python also graph GPS and all other sensor data, then I start tuning EKF when When I get the Expected result I stop, In that way I think I get my optimal Q and R matrices values? am I right? please guide me!
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.