In 2008 Seattle based engineer William Beaty in a series of experiments established a so-called Traffic Waves Theory. The underlying idea of that theory was that by knowing how traffic waves are formed, drivers could sometimes reduce their effect by increasing the distance to the car immediately ahead of them. This leads to breaking less often, and ultimately, decrease the accumulation of vehicles/traffic around them.

Later on, it was also proved in a series of experiments that even intelligent cruise control systems, could be used to clear up congestion.

But to motivate drivers for the described behavior, we as humans need to unlearn everything we were taught about standing in lines.

What it does

The system we introduce allows modern self-driving vehicles to purchase routes, negotiate the road space and pay for the rights of way.

It will allow to Financially incentivize both Human and Autonomous driving vehicles to maintain a good driving behavior by getting paid for the road space and yielding to the other cars. The system will also allow the purchasing of the routes for autonomous vehicles.

We hope that this system will evolve to become a part of our daily traffic infrastructure.

How we built it

To participate in the Mobi Grand Challenge, we’ve partnered with our friends at Decentralized Tech – a San Francisco startup behind the decentralized platform that helps companies to monetize robots-as-a-service.

For our demo project, we have utilized advanced developments in the Robot Operating System as well as Robonomics Platform, Ethereum infrastructure for cyber-physical systems' integration into Smart Cities and Industry 4.0

Challenges we ran into

The main challenge was to make transactions faster and cheaper. To solve this challenge we have utilized Robonomics Platform with separated on-chain and off-chain communication. We have also set up an Ethereum Sidechain with the mainnet token swaps.

Also, for speeding up the transactions, we are not waiting for the contract execution to exchange signed messages, which miners could add to blockchain later. With this signed messages we don't need to wait for the transaction to be included to the block.

Accomplishments that we're proud of

We have developed a real-life PoC for a peer-to-peer marketplace for Machine-to-Machine services.

The Infrastructure of such market can be installed even on EDGE devices like 5G routers.

We have also introduced a very impressive Traffic Management System with smart contract-based routing permissions.

What we learned

Decentralized technologies for both Self-Driving Cars and Drones have very similar technical tasks and obstacles.

Peer-to-peer technology we have built will save bandwidth by allowing cars to communicate directly without accessing the internet.

What's next

In the next stage of the project, we are planning to introduce web GUI for the general public to be able to interact with the system and track the vehicles on the map.

The future stages of development will also include experiments with different Blockchain types and improvements in speed and robustness of the system.

The final stage of the project will include switching from virtual simulations and Duckietown to real life autonomous cars. A real-life test of Vehicles with Level 3,4 autonomy.

Built With

  • aira
  • duckietown
  • ethereum
  • gazebo
  • haskell
  • nix
  • python
  • robonomics-platform
  • ros
  • solidity
Share this project: