AI for Greener Cities Challenge
Harvest driving video data from urban areas to get insight into the state of the environment of our cities. Use this data to come up with solutions to improve the sustainability of a city.
What could you do for sustainable cities, if you had a fleet of trucks with cameras driving around the town?
Challenge completed. Watch the recording of the AI for Greener Cities final results presentations to discover that the teams came up with:
About the Challenge
In 2021 4,153 parcels were sent worldwide every second and this number is expected to double by 2026. European countries agreed to achieve a 55% reduction in emissions by 2030. One of the ways to achieve this ambitious goal is to reduce emissions from delivery trucks by 50%. For this purpose, DPD is aiming to deploy only vehicles that emit very little or no CO2 in Europe’s 225 largest cities by 2025. The company started to equip their cars with sensors and record their drives gathering valuable data to innovate in that direction. It created an unexpected opportunity - giving us more insight into our cities and offering solutions on how to make them more sustainable.
For this challenge, FruitPunch AI teamed up with DPD and Jheronimus Academy of Data Science to find out the potential of driving video data to improve the sustainability of urban areas. Delivery trucks can help identify areas with lots of street waste, or neighborhoods that could use more vegetation for example. You are free to come up with your own game plan. You’ll be supported by data science experts from FruitPunch AI, JADS and DPD.
Our challenge partners
GOAL: Analyze driving datasets and come up with an AI-powered solution for a more sustainable city
- Use Data Engineering on large video driving datasets (~2TB)
- Apply techniques like object detection and segmentation to sensor data from urban environment
- Map the environment for things like vegetation coverage, pollution hotspots where cyclists can't drive or busy traffic spots that can cause bad air quality.
- Write a project plan for your idea to improve sustainability in cities.
We’ll be zooming into the newest methods in computer vision for:
- Object detection
- Edge computing
- Image segmentation
- Transfer learning
Valuable results will be implemented on the DPD delivery vehicles improving the sustainability in 68 cities throughout Europe.
Based on open source datasets, the outcomes of the challenge will be reproducible. We’ll share the results via our channels.
Who are we looking for?
We need creative data science and machine learning minds with experience in computer vision to come up with a game plan. You'll collaborate with a diverse team of up to 50 international data specialists in subteams, all tackling this problem from different angles.
You can join as a contributor (~12 hours per week commitment for 10 weeks) or coach (2-4 hours per week, only for experienced ML professionals). We’ll organize a masterclass on relevant topic during the challenge to bring you up to speed.
Did you know
🌍 The world’s cities occupy just 3% of land, but account for 60-80% of energy consumption and 75% of carbon emissions.
🇪🇺 EU countries want to achieve 55% reduction in emissions by 2030.
🚚 One of the ways is to reduce emissions from delivery trucks by 50%.
📦 In 2021 4,153 Packages were sent every second. This number is expected to double by 2026.
🌆 DPD operates more than 8,000 vehicles from 68 locations in Europe and delivers over 250 million parcels a year.
Application Deadline: 12 June 2022
Challenge Kick-off: 13 June 2022
Midterm Presentations: 11 July 2022
Final Presentations: 22 August 2022