Wolfgang Stefani's repositories
Vehicle-Localization
High-performance localization software for autonomous vehicles. A particle filter is combined with a map to localize a vehicle.
Traffic-Sign-Classifier
This program uses a deep neural network with several convolutional layers to classify traffic signs. The model is able to recognize traffic signs with an accuracy of 96,2%. It was trained and validated using the German Traffic Sign Dataset with 43 classes (types of traffic signs) and more than 50,000 images in total.
Highway-Path-Planning
Path planner that creates smooth, safe trajectories for an autonomous vehicle to follow along a 3 lane highway with traffic. The path planner is able to keep inside its lane, avoid hitting other cars, and pass slower moving traffic all by using localization, sensor fusion, and map data.
PID-Controller
Control algorithm to control steering angle to navigate a vehicle around a track
Self-Driving-Car_System-Integration
Using ROS to program a self-driving car
Extended-Kalman-Filter
This repository implements an Extended Kalman Filter (EKF) in C++. A Kalman Filter can be utilized to estimate the state of a moving object of interest with noisy lidar and radar measurements. In this project a bicycle is detected that travels around our vehicle. The EKF and the measurements are used to track the bicycle’s position and velocity.
Advanced_LaneLines_Detection
In this project, a software pipeline was written to identify the lane boundaries in a video from a front-facing camera on a car using computer vision techniques with openCV.
Behavioral-Cloning
This repository contains files for cloning a human driver's behavior and training an autonomous vehicle to imitate the driving behaviour. Deep neural networks and convolutional neural networks are used to clone the driving behavior. With input from camera data, it will output steering angles for an autonomous vehicle. The model is trained, validated and tested using Keras.
udacity-traffic-light-classifier
A classifier that takes in images of traffic lights and outputs a label that classifies the image as red, green or yellow using machine learning and computer vision algorithms
CarND-LaneLines-P1
When we drive, we use our eyes to decide where to go. Naturally, one of the first things we would like to do in developing a self-driving car is to automatically detect lane lines using an algorithm. In this project we will detect lane lines in images using Python and OpenCV, which is a package that has many useful tools for analyzing images.
CarND-Path-Planning
Udacity Self Driving program Path planning project. Using Sensor fusion data to program highway lane changing behaviors
computer-vision_canny_edge_detection
With this computer vision technique, the canny edge detection algorithm, it is possible to detect the shape of objects in pictures. This can be used for example lane line detection or detection of other road users like vehicles, bicycles, pedestrians, etc.
computer-vision_color-selection
Several parts of an image can be blacked out according to RGB thresholds and the result are images with only the selected color. This is useful for lane detection where for example only white colors pixels are retained.
Route-Planner
This repository includes all project files from Udacity's Nanodegree Intro to Self-Driving Cars Project 4
udacity-reconstructing-trajectories-from-sensor-data
This repository contains solution code for the Udacity project "Reconstructing Trajectories from sensor data" in vehicle motion and control