Traffic Flow Optimisation using the D-Wave Quantum Annealer. Qubits Europe, Munich Dr. Christian Seidel

Transcription

1 Traffic Flow Optimisation using the D-Wave Quantum Annealer Qubits Europe, Munich Dr. Christian Seidel

2 Christian Seidel Lead Quantum Computer Scientist, Volkswagen Data:Lab Joined the Volkswagen Data:Lab at its beginning in 2014 Areas of work since then: Connected Car, Future mobility, Smart City Internet of Things Natural Language Processing, Semantic Analysis, Quantum Computing 2

3 Team Christian Seidel Volkswagen Data:Lab, Munich Florian Neukart Volkswagen Group of America Code:Lab, San Francisco Patrick van der Smagt Volkswagen Data:Lab, Munich David von Dollen Volkswagen Group of America Code:Lab, San Francisco Isabella Galter Volkswagen Data:Lab, Munich Andrea Skolik Volkswagen Data:Lab, Munich Michael Streif Volkswagen Data:Lab, Munich 3

4 Quantum Computing just a hype or a real thing? 4

5 The Question that drove us Is there a real-world problem that could be addressed with a Quantum Computer? 5

6 YES: Traffic flow optimisation Everybody knows traffic (jam) and normally nobody likes it. Image courtesy of think4photop at FreeDigitalPhotos.net 6

7 Public data set: T-Drive trajectory Beijing ~ Taxis data example: 7

8 D-Wave calculation model Quadratic Unconstraint Binary Optimisation (QUBO) During the quantum annealing process the system evolves to the lowest energy level. This requires your problem to be formulated as an Ising Model: or as a QUBO: 8

9 Transforming the real world problem for the Quantum Computer Example: Simplified graph structure representing a route grid. 2 cars with 3 route options on a 2 x 2 grid. Car Route Binary variable Car 1 #1: s0,s3,s6,s9 Q 11 Car 1 #2: s0,s3,s8,s11 Q 12 Car 1 #3: s2,s7,s10,s11 Q 13 Car 2 #1: s0,s3,s6,s9 Q 21 Car 2 #2: s0,s3,s8,s11 Q 22 Car 2 #3: s2,s7,s10,s11 Q 23 9

10 Creating the cost function for each route segment More cars on one street lead to higher costs Street segment Associated cost function Value Example: s0 (Q 11 + Q 12 + Q 21 + Q 22 ) 2 4 s3 (Q 11 + Q 12 + Q 21 + Q 22 ) 2 4 s6 (Q 11 + Q 21 ) 2 1 s9 (Q 11 + Q 21 ) 2 1 s8 (Q 12 + Q 22 ) 2 1 s11 (Q 12 + Q 22 + Q 13 + Q 23 ) 2 1 s2 (Q 13 + Q 23 ) 2 0 s7 (Q 13 + Q 23 ) 2 0 s10 (Q 13 + Q 23 ) 2 0 (Q 11 + Q 12 + Q 21 + Q 22 ) 2 + (Q 11 + Q 12 + Q 21 + Q 22 ) 2 + (Q 11 + Q 21 ) 2 + (Q 11 + Q 21 ) 2 + (Q 12 + Q 22 ) 2 + (Q 12 + Q 22 + Q 13 + Q 23 ) 2 + (Q 13 + Q 23 ) 2 + (Q 13 + Q 23 ) 2 + (Q 13 + Q 23 ) 2 = 12 Goal: minimise the all-over-costs => distribute cars to different streets 10

11 Data preprocessing Transforming the geo-coordinats to street segments using OSMnx, a Python package for street networks Getting real / valid alternative routes via HERE-maps requests. 11

12 Beijing Traffic Heatmap Traffic in the city cars Detail: route to the Airport 418 cars We assigned each of the 418 cars 3 possible routes to reach the airport Size of the problem space: 3^418 12

13 Code snippet and output 13

14 Result: un-optimised vs optimised traffic 14

15 Further Improvements Due to the 2,5 months project time we threated in this test all street equally. This is obviously a simplification. Additional constraints could be: Street capacity (highway vs alley) Residential zone Data set improvements: Frequent updates to react on constantly changing traffic situations (other data set) more cars etc 15

16 Publications Traffic flow optimization using a quantum annealer Quantum-enhanced reinforcement learning for finite-episode games with discrete state spaces Quantum-assisted cluster analysis 16

17 Follow up projects example Machine Learning Recommendation System Use output for Clustering/Classification 17

18 And a lot more Material Simulation: talk held by Michael Streif Experience with using the D-Wave: talk held by Isabella Galter University cooperation: QASAR - Results and hands-on demonstration talk held by Sebastian Feld + Thomas Gabor of a joint project of Volkswagen and LMU Investigating the annealing path talk held by Kristel Michielsen project with the Forschungszentrum Jülich 18

19 Lessons learnd There are sooo many projects for a Quantum Annealer out there The D-Wave Quantum Annealer can help solving real world problems (Prototype was done in 2,5 months) transforming the real world problem into a QUBO takes the most time Problems, larger than the chip s capacity can be solved by decomposition using a hybrid solver (i.e. QSage, QBsolve) Due to the chimera graph structure of the quantum chip: The chip is not fully connected, so Qubit chains need to be created Challenging the Precision: find the right values for chain strengths (Qubit connection) and penalty weights Late Easter Wish: We d like to have a Java API for the D-Wave 19

20 Questions? 20