search

a-b-street / abstreet

Transportation planning and traffic simulation software for creating cities friendlier to walking, biking, and public transit
https://a-b-street.github.io/docs/
Apache License 2.0
7.66k stars 340 forks source link

Support for traffic data by OpenDataCam #228

Closed jvolker closed 3 years ago

jvolker commented 4 years ago

https://github.com/opendatacam/opendatacam

OpenDataCam is an open source tool to quantify the world. It quantifies and tracks moving objects with live video analysis. It is designed to be an accessible, affordable and open-source solution to better understand interactions in urban environments.

this entire setup is priced around $150. All software is based on open source components and runs completely locally. The software features a friendly user interface and is currently optimised for detecting and counting traffic participants

A group at CityLAB Berlin is currently evaluating this system for traffic analysis. Could this help to provide realistic traffic data when focussing on one particular crossing within A/B street?

What would be necessary to import and use the data of the camera within A/B street? Counter history API: https://opendatacam.github.io/opendatacam/apidoc/#api-Recording-Counter_history_(CSV)

Thanks.

dabreegster commented 4 years ago

I can think of two possible uses.

1) Validating a traffic model. If we come up with a travel demand model, we could run the simulation and then use the throughput tool on a lane or intersection and see how many different agents pass through at some time. You could then take the same measurements in real life and compare counts. I don't know how to make use of the results. If they match, is it a sign to trust the simulation? If they don't, what action could we take to improve it?

2) Generating trip demand input. If you have counts around many areas, I could imagine some kind of graph-based algorithm that tries to figure out exact trip start and end-points that roughly match the measured counts along different roads. I don't have any knowledge of this algorithm (beyond a vague memory of somebody talking about this approach), and I'm not promising I could implement it, or that the results would be at all realistic.

dabreegster commented 3 years ago

Just cleaning up old issues. I'm still not sure how to make use of this kind of data. We're still barely touching the problem of generating travel demand, except for #424. If you have any ideas how to incorporate it, please reopen.