Traffic Man

Full Presentation

https://www.youtube.com/watch?v=r6WmgEDuyKk

Inspiration

Large cities have a lot of congestion and sitting in traffic is rough, estimates put the loss of productivity while sitting in traffic in the United States >$70 billion. Optimizing traffic light policy is one step closer solving the issue is congestion particularly in large cities where the most congestion occurs, and a majority of the economic loss.

What it does

Actuated traffic signal policy optimization using automated reasoning. The program simply takes in constraints which would cause conflict and handles traffic light transition, hypothetically this creates a pretty good model as we're able to reason about when to change lights.

How I built it

The logical backbone is written using s(CASP), the api, and fronted are both written using python.

Challenges I ran into

s(CASP) :- not prolog.

Accomplishments that I'm proud of

Getting a baseline extensible model working in s(CASP), extending s(CASP) to more real world use case with implementation of api.

What I learned

s(CASP) would benefit greatly from a main line prolog release as a module rather than a command line tool!

What's next for Traffic Man

You can add more constraints, and build off this baseline model for more varied use case, github linked below. How to go about doing this is by adding constraints to the conflict(X) rule.

Built With

• python
• s(casp)