|
Hi all, I am using kalman filters for localisation of a robot. I am trying to implement the propagation and updation equations of a kalman filter shown below, using MATLAB. xhat(k+1) = phi * xhat(k) where phi is a state transition matrix. xhat(k) = a state vector that have the following values left wheel velocity, right wheel velocity and orientation theta as a 3 x 1 matrix. where phi = [1 0 0; 0 1 0; -.75 .75 0]. P(k+1) = phi*p(k)*phi(tranpose) + Q where Q is a system noise covariance matrix. P is a covariance matrix. My problem is I am a beginner and not sure of how exactly the above equation can be realized as a matrix equations/expressions. i have tried implementing the above equation by initialising xhat as a null matrix and later on finding/calculating the elements of this matrix as a for loop. But I think the errors i am getting is related to inappropriate method of realising it as a matrix equation without violating the rules and thus avoiding errors in matlab. So could some one give me a basic idea of how it can be done either in a command line form or as a function. Regards Suresh |
|
|
Matlab implementation for kalman filters
Started by ●December 18, 2002
Reply by ●December 18, 20022002-12-18
|
Setting the system up with a NULL is not generally appropriate. If I remember my mathematics correctly ... you will need to define some boundary conditions to initialize the system. Regards, Tom Dillon -----Original Message----- From: suresh <> [mailto:] Sent: Wednesday, December 18, 2002 9:37 AM To: Subject: [matlab] Matlab implementation for kalman filters Hi all, I am using kalman filters for localisation of a robot. I am trying to implement the propagation and updation equations of a kalman filter shown below, using MATLAB. xhat(k+1) = phi * xhat(k) where phi is a state transition matrix. xhat(k) = a state vector that have the following values left wheel velocity, right wheel velocity and orientation theta as a 3 x 1 matrix. where phi = [1 0 0; 0 1 0; -.75 .75 0]. P(k+1) = phi*p(k)*phi(tranpose) + Q where Q is a system noise covariance matrix. P is a covariance matrix. My problem is I am a beginner and not sure of how exactly the above equation can be realized as a matrix equations/expressions. i have tried implementing the above equation by initialising xhat as a null matrix and later on finding/calculating the elements of this matrix as a for loop. But I think the errors i am getting is related to inappropriate method of realising it as a matrix equation without violating the rules and thus avoiding errors in matlab. So could some one give me a basic idea of how it can be done either in a command line form or as a function. Regards Suresh _____________________________________ Note: If you do a simple "reply" with your email client, only the author of this message will receive your answer. You need to do a "reply all" if you want your answer to be distributed to the entire group. _____________________________________ About this discussion group: To Join: To Post: To Leave: Archives: http://www.yahoogroups.com/group/matlab More DSP-Related Groups: http://www.dsprelated.com/groups.php3 ">http://docs.yahoo.com/info/terms/ |
Reply by ●December 19, 20022002-12-19
|
Hi Suresh I dont know whether the formulation of the problem is correct or not. But to get help about the MATLAB error messages, you should post the MATLAB code you have tried and the equations you are trying to implement along with the error messages you are getting. It is very easy to learn and start coding m scipts. Navan --- "suresh <>" <> wrote: > Hi all, > > I am using kalman filters for localisation of a > robot. > > I am trying to implement the propagation and > updation equations of > a > kalman filter shown below, using MATLAB. > > xhat(k+1) = phi * xhat(k) > > where phi is a state transition matrix. > > xhat(k) = a state vector that have the following > values > > left wheel velocity, right wheel velocity > and orientation > theta as a 3 x 1 matrix. > > where phi = [1 0 0; 0 1 0; -.75 .75 0]. > > P(k+1) = phi*p(k)*phi(tranpose) + Q > > where Q is a system noise covariance matrix. > P is a covariance matrix. > > My problem is I am a beginner and not sure of how > exactly the above > equation can be realized as a matrix > equations/expressions. i have > tried implementing the above equation by > initialising xhat as a > null > matrix and later on finding/calculating the > elements of this matrix > as a for loop. But I think the errors i am getting > is related to > inappropriate method of realising it as a matrix > equation without > violating the rules and thus avoiding errors in > matlab. So could > some > one give me a basic idea of how it can be done > either in a command > line form or as a function. > > Regards > Suresh __________________________________________________ |
