Human-Robot Interaction for a World Without Plastic Litter

The BeachBot detects small litter in outdoor areas, like beaches and parks and needs your help to detect even more.  It is programmed with artificial intelligence enabling it to efficiently map and collect litter, detecting and acting at the same time.  A swarm of robots can collaboratively hunt for litter.

The project incorporates a human-robot interaction. By using a gaming application, anyone with a smartphone can contribute to improve the litter detection algorithm and make the robot smarter. The process is simple: The robot scans the area and collects litter images. Each raw image is then added to a database which forms the input for a labeling game where the gamer can match images with the right type of litter. This way, the initial detection is verified by a reliable (=human) source. This is a win-win formula since (A) the detection algorithm will be improved enabling the robots to perform their task autonomously over time and (B) increasing human awareness on the effects of littering will contribute towards a changing mindset and reduce generating waste at the source. 

Edwin Bos likes the nature of the Dutch coast. What he does not like is litter at the beaches. So he started to think and act.  He and his fellow entrepreneur Martijn Lukaart have built the BeachBot together with the passionate people at the Technical University Delft. Edwin and Lukaart are the co-founders of TechTics, a consultancy based in The Hague that works to resolve social issues with technology.