B7 Informatics 2 WS 2021/22

Website of Prof. Dr. Barne Kleinen, Professor for Media Informatics (Bachelor/Master) at HTW Berlin
HTW Berlin
Programming and Web Technology @ IMI

Info2: Exercise 10: Getting from A to B

     <prev next>

S-Bahn-Ring S-Bahn Ring generated with https://github.com/bkleinen/bvg-graph based on OpenStreetMap

Pre-Lab

P1. Define an interface data type for a weighted graph. What methods does it need? What are the signatures?

P2. Read up on Depth-First-Search to compute a path in a given graph. Sketch the algorithm on paper. Do you have an idea how you could find the shortest path, instead of just a path?

P3. Read up un Dijkstra Algorithm to compute the shortest path in a given graph. Sketch the algorithm on paper.

P4. Your algorithm will probably need an adjacency matrix or an adjacency list as its data structure. Think about how you would implement such a structure, if you only had linked lists available. What methods will you need for your data structure?

Assignment

Read through everything first and think about who will do what. Our goal is to write a program to determine how to get from A to B, either fast or cheap. We first need some test data.

  1. Design and implement a data type WeightedGraph that uses either an adjacency list or an adjacency matrix.
  2. While one partner is doing this, the other one should write a class that reads a graph from a file. See notes on the file format and the example file below!
  3. Meanwhile, your partner writes a method that takes a graph, picks two vertices at random, and finds the shortest path, that is, the one with the least travelling time. Make a method to print out the path in a readable format.
  4. Starting from S Schöneweide Bhf (Berlin) compute the shortest travel times to the 4 Stations below.

Your Dijkstra implementations should yield the following travel times:

(Which is a plausible result given that the graph doesn’t consider time spent in atations.)

For the bored

  1. Use your data structure to print out all the vertices n steps from a given vertex.
  2. List the travel times to all stations from S Schöneweide.
  3. Are there multiple minimal paths? Print them all (this is very tricky!).

Graph Example and Test Data

Test Data

graph1.txt contains a small example graph to test both algorithms. For the Dijkstra Algorithm, result1.txt contains the cheapest path costs if you start at vertex 1.

BVG U+S Bahn

The graph data in bvg.txt contains a simple extract of the Berlin U+S Map and has the following format:

<from-vertex> <to-vertex2>,<weight2> <to-vertex2>,<weight2>

That is, each line represents a vertex with all its outgoing edges, e.g.

060049202852 060049201862,90 060066101852,108

means that there is an edge from 060049202852 to 060049201862 with weight 90, that is, the travel time from

060049202852, S Sundgauer Str. (Berlin) to 060049201862, S Zehlendorf (Berlin) is 90 sec.

You find the station names in stations.txt

The graph data was extracted from the GTFS-Data provided at https://www.vbb.de/de/article/fahrplan/webservices/datensaetze/1186967.html using the rails app https://github.com/bkleinen/bvg-graph

Update WS 20/21: I’ve found the current data at https://daten.berlin.de/datensaetze/vbb-fahrplandaten-gtfs, you can read about the GTFS format on Wikipedia or on the GTFS für Deutschlandsite.

Lab Report / What to turn in

All info on the lab reports can be found on the Labs page.

Also answer the following questions in your report:

  • Ex. 1: How are you going to store the weights?
  • Ex. 3: What class will these methods belong to?
  • Do your implementations return the correct results?