MATLAB openDRIVE Tool

Introduction

This repository provides a MATLAB tool for the usage of the [openDRIVE] format. The tool is parsing the xml/xodr file into a MATLAB class structure which is based on the openDRIVE specifications. Additionally, the map is enriched with additional attributes and methods to enable fast queries and easy working with the map. Hence, the map can be loaded and used within MATLAB.

The tool was first used and mentioned in the work "An Entropy Based Outlier Score and its Application to Novelty Detection for Road Infrastructure Images" published in 31st IEEE Intelligent Vehicles Symposium 2020 [1].

In this release, the main functionalities are

parsing the file into the class structure
enrich the map for a given Region of Interest ROI
plot the enriched map for a given ROI
get map information for a given 2-D point

Note:

So far, the map points are only constructed in 2-D (xy).
No objects are considered.
tPoints are not consistent to the t variable as used in openDRIVE. Here, it represents the angle of the tangent at a certain s.

Examples


Plot crossing	Plot complete view of an example

Usage

The basic usage of the class can be found in main.m. The main steps are as follows:

Construct an empty map as:
- MAP = ODR.OpenDRIVE()
Parse the xml/xodr file as:
- MAP = MAP.OpenDRIVEXML('filename')
Enrich the map for a ROI as:
- MAP = MAP.enrich(ROI)
  - Specific area: ROI = [x_min x_max y_min y_max]
  - Whole map: ROI = []

The above listed steps are required for the following methods to work.

Plotting the enriched map:
- MAP.plot([],ROI,[])
  - Every lane of type 'driving', 'sidewalk', 'restricted', which got constructed during enrich and all signals will be plotted. To add more lane types to be plotted, they need to be defined in plot.m.
  - The input arguments are:
    1. ignoreStrings: Cellarray containing strings of elements which should not be plotted. Example:
      - {'signals','sidewalk'}
    2. ROI sets the axis limits. [] will automatically set the axis to contain the full map.
    3. orientation set the orientation of the map in radians, with [] no orientation (upper is north). Note: Since the graphic objects themselfs are rotated, the axis won't be correct after rotation. Use this property carefully.
  - If the signs of the signals shall be plotted, a folder containing the corresponding .png files has to be added to the MATLAB path. The filenames have to be the corresponding sign ID.
Determine the map information for a point
- [laneSectionNr, laneID,roadID] = MAP.getRoad_pos(point)
  - Given the MAP this method determines all roadID, laneID, laneSectionNr where a point lies on. The output variables are all of the same shape. Example:
    - roadID = [4;8], laneSectionNr = [1;1], laneID = [-1;1]
    - The first possible combination is: The point is located on road 4 on the lane section with nr 1 on the lane with the id -1. Second case respectively.

Class and Packages Structure

The class structure is embedded into a package foldering. The structure will be briefly explained for a single case, but it holds for all the classes.

All classes, which are related to openDRIVE are within the ODR package +ODR. Other, general functions are located in +Utils. The class defintion and the methods of openDRIVE can be found in the folder +ODR/@OpenDRIVE. The file, which has the name of the folder contains the class defintion, the remaining files are methods of this class. A nested structure is deployed, hence the defintions for example the lane class are located in +ODR/+roadPKG/+lanesPKG/+laneSectionPKG/@lane.

Data and Example Files

To test the provided tool, one may download the sample files from http://www.opendrive.org/download.html and place them in the same folder as the project. The filename (without .xodr) has to be defined in the main.

License

LICENSE

Reference

[1]: An Entropy Based Outlier Score and its Application to Novelty Detection for Road Infrastructure Images; Jonas Wurst, Alberto Flores Fernández, Michael Botsch and Wolfgang Utschick; 2020 IEEE Intelligent Vehicles Symposium (IV); Paper
[openDRIVE]: Most recent specification can be found in http://www.opendrive.org/download.html

Citation

To cite the work use:

@INPROCEEDINGS{WurstIV20,
author={J. {Wurst} and A. {Flores Fernández} and M. {Botsch} and W. {Utschick}},
booktitle={2020 IEEE Intelligent Vehicles Symposium (IV)},
title={{An Entropy Based Outlier Score and its Application to Novelty Detection for Road Infrastructure Images}},
year={2020}
}

JWTHI / openDRIVE-Matlab