Self-Driving Car Engineer Nanodegree Program
Utilize an Unscented Kalman Filter to estimate the state of a moving object of interest with noisy lidar and radar measurements.
- UKF algorithm that predicts object position to the current timestep and then updates the prediction using the sensor output;
- UKF algorithm implemented with measurement functions for Lidar vs. radar, and calls the function according to the sensor type;
- RMSE less than
$[.09, .10, .40, .30]$ as shown in the figure below: