Sonic

SONIC TIER1 GROUP NUMBER 100 BOUNTIES: MACHINE LEARNING, MOST INNOVATIVE IDEA

Inspiration

Greenhouse gas (GHG) emissions from transportation account for about 30 percent of total U.S. greenhouse gas emissions, making it the largest contributor of GHG emissions and a cause of concern. Between 1990 and 2019, GHG emissions in the transportation sector increased more than any other sector. Comparing the efficiency of different modes of transportation, optimizing or analyzing infrastructure for a specific type of sustainable transportation, or quantifying the impact of adopting a specific type of transportation as the status quo.

Machine learning application was inspired by "Captcha" where asked to identify cars/bikes/hills.

Are Street Curbs The Next Frontier In Sustainable Transportation? Our streets are more congested than ever with benefits to having LESS cars on the road include the following:

• Less congestion and fewer emissions with less traffic
• Safer streets for pedestrians with enforcement of tickets
• More business for restaurants and retailers
• Dedicated curb access for commercial operators
• More efficient and productive curbs for all
• Increased revenue from parking/(un)loading 

What it does

SONIC stands for Sustainable Optimization Navigation Investigating CO2 and applies computer vision to curb and street management to optimize sustainable transportation.

USER CONSUMERS: Parking can be a complete nightmare. Drivers gain more dedicated access to the curb to ensure you always have the safest and quickest way to pick up and drop off while reducing traffic flow, pollution and congestion.

Al-POWERED: Get real-time availability data minimize time spent searching for parking reducing driver stress

PAYMENTS: Easy and seamless payment through automated invoicing car operators by the minute - align policies and pricing with multimodal demand

SEAMLESS: Reduce wait time for pick up and drop off locations - sends information about the size of available spots so drivers can tell whether their vehicles will fit.

GO GREEN: Supports urban mobility allows drivers to spend less time and fuel aimlessly driving around the block trying to find a space. 

USERS BUSINESSES: Access real-time curb and street activity across freight, ride hailing, delivery, and passenger vehicles, and buses, e-scooters, bikes, and pedestrians. Automatically alert enforcement officers of violations in real time and send citations directly to vehicle owners. Check curb availability through a web app into any freight, ride hailing, or food delivery vehicle. Make payment for time spent at the curb seamless for all users, including freight, ride hailing, and food delivery vehicles.

BUSINESS: Align your policies and pricing with multimodal demand and increase revenue from curbside parking, loading, and unloading

PROFITS: Serve more customers through increased parking turnover - guest parking, pickup experience and food delivery drivers

ENFORCEMENT: Identify violations in real time and automatically detect, notify, and enforce common infractions

SUSTAINABLE: Help companies to measure their CO2 emissions to comply with regulation, and cities to reduce air pollution 

How we built it

Computer vision devices installed in windows of participating businesses record when curb space is occupied near the shop and relay that information to the drivers. AI-powered technology helps cities, airports, fleets, and businesses increase revenue with access to real-time curb and street activity across freight, ride hailing, delivery, and passenger vehicles, and buses, e-scooters, bikes, and pedestrians

Multiple object detection using pre trained model in TensorFlow.js was used for video analytics of cars from https://blog.tensorflow.org/2021/01/custom-object-detection-in-browser.html in addition to a recommender system for vehicle selection from https://ieeexplore.ieee.org/abstract/document/8811083

Comparing the efficiency of different modes of transportation allows Sonic to identify car types from curbside videos and share live parking data with drivers and administrators and gain insights and monitors parking and traffic for every mode of transit:

Built With

• computer
• javascript
• not-microsoft-office
• tensorflow
• vision