This project contains an implementation of Extended Kalman Filters for the Udacity Nanodegree program.
Dataset #1 | Dataset #2 |
---|---|
![]() |
![]() |
video 1 | video 2 |
Once the install for uWebSocketIO is complete:
mkdir build
cd build
cmake ..
- sometimes you may want to specify compilers manually for example :
cmake -D CMAKE_C_COMPILER=clang -D CMAKE_CXX_COMPILER=clang++ ..
- sometimes you may want to specify compilers manually for example :
make
./ExtendedKF
This implementation meets accuracy requirement: RMSE <= [.11, .11, 0.52, 0.52]
. It has higher values at the beginning, but it meets the requirements after few rounds.
- Implementation follows defined steps for the Kalman filter.
- The algorithm uses the first measurements to initialize the state vectors and covariance matrices.
- Upon receiving a measurement after the first, the algorithm predicts the object position to the current timestep and then update the prediction using the new measurement.
- The algorithm sets up the appropriate matrices given the type of measurement and calls the correct measurement function for a given sensor type.