Skip to content

zoespot/CarND-Kidnapped-Vehicle-Project

 
 

Repository files navigation

Kidnapped Vehicle Project

Self-Driving Car Engineer Nanodegree Program Term2 project3

My code built without error with cmake .. && make, and connected to GUI with ./run.sh. It runs with success. Success snapshot

The car has been kidnapped on the map with (noisy) GPS estimate as its initial location, and lots of (noisy) sensor observations and control data. The key steps include initializing particles with GPS data, predicting with velocity and yaw rate and update particles weight and resampling. 100 particles are chosen to balance efficiency and accurancy.

The steps of weight updates for each particle start with observation points transformation to map coordinates, then finding the nearest landmark for each observation point, and finally multiplied the weight by the multivariate Gaussian probability measuring the distance between the observation points to its associated landmark.

About

Kidnapped Vehicle Project for Udacity Self-Driving Car Nanodegree

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 99.4%
  • Other 0.6%