Inspiration

Every year, around 8 million tonnes of plastic ends up in our oceans. By 2025, the number of plastics in the ocean is predicted to double. Currently, according to researchers 245 million tonnes of plastic is floating on the ocean’s surface, but heavier plastics have sunk to the seafloor, where it is impossible to catalogue or quantify. Ocean plastic is dissolving into microplastics, which insidiously work their way into the bodies of fish, birds, and even humans.

To date, plastic debris in rivers has only been monitored in a handful of rivers. Through affordable and accurate automated measurement methods, such as cameras, UAVs and spaceborne remote sensing, global coverage of riverine plastic debris monitoring is within reach in the coming years.

Whether by autonomous drones, remote-controlled robots or swarms of conveyor belts, it’s clear that something drastic needs to happen to clean up the world’s oceans. We’re currently on track to have more plastic than fish in the ocean by 2050, at which point it will be too late to save the oceans, and ourselves. We believe that if we don’t take responsibility for our environment and fix the prevailing issues, we might become the reason for life extinction on our planet.

What it does

We propose to design AQUABOT, an intelligent floating drone, embedded with a 4k camera above it and a filtering mesh underneath.

The floating drone will move on the water with help of batteries and propellers detecting and collecting all floating waste (plastic, wood, paper cups, textile waste, nets) at the same time sending real-time data from embedded sensors regarding the water quality.

1.Detecting and collecting floating waste. 2.Detecting the obstacle. 3.Collecting Real-time data of water. 4.Categorizing the collected waste. 5.Remote control. 6.Adaptability.

How we built it

We propose to design AQUABOT, an intelligent floating drone, embedded with a 4k camera above it and a filtering mesh underneath.

The floating drone will move on the water with help of batteries and propellers detecting and collecting all floating waste (plastic, wood, paper cups, textile waste, nets) at the same time sending real-time data from embedded sensors regarding the water quality.

Detecting and collecting floating waste

We propose to use an underwater proximity sensor along with a 4k embedded camera which will be enabled above the drone to detect the waste and its type (plastic, wood, paper cups, textile waste, nets etc.).

Also, categorizes them as recyclable and non-recyclable waste, along with percentage analysis which will help in waste management.

Detecting the obstacle

By analysing the real-time video stream, which will be acquired through the embedded camera, using object detection technique. This technique which will be implemented through Deep learning will safeguard the drone from collisions with surroundings like rocks, moving boats, river bed, etc.

Collecting Real-time data

By using underwater proximity sensor, hydrostatic pressure sensors, pH sensor, salinity sensor and Fluorescent Chemosensors to detect pH value of the water, depth of the river, amount of nitrates, ammonium salts, reactive oxygen species, Biothiol, metal ions, and toxic gases.

This data will be sent to the Central unit using a Microcontroller which will be embedded in the drone (Raspberry pi 4, NVidia Jetson).

Categorizing the waste

By using the Object detection technique implemented in deep learning, the drone will detect various types of trash present on the water surface.

This data will provide information about the types of waste, which will help in waste management of the procured waste.

Remote control

The proposed drone can act as an autonomous drone and also be controlled with an IR wireless remote control.

Adaptability

The drone is re-programmable according to the need which will help to add various functionalities, without major changes in the hardware design. This makes the proposed solution custom-made, cost-efficient and reusable, in comparison with the existing models in the market.

On-board sensors: 4k Camera, Underwater proximity sensor, Temperature sensor, pH sensor, salinity sensor and Fluorescent Chemosensors

Waste type: Plastics, floating debris

Operation area: Semi-confined water surfaces (Rivers, canals, lakes)

Steering control: Remote control and autonomous

Challenges we ran into

The major challenges we faced are:

  1. How to make the drone power-optimized, drone shouldn't move randomly on the water. then we came up with object detection that provides drones with the intelligence to move along the direction where wastage is present.

  2. How to make a drone withstand tides of water and wind.

Accomplishments that we're proud of

  1. making drone move on its intelligence.
  2. contributing to restoring the ecosystem.

What we learned

  1. how humans are keeping a blind eye on the plastic they use and innocent marine life is facing a lot of damage for which they are not even responsible. 2.learnt deep learning techniques for object detection.
  2. how to integrate all sensors at once.

What's next for Untitled

Share this project:

Updates