HOW WILL WE SAVE THE RAINFORESTS? WE HAVE A BOLD PLAN USING A.I. AND BLOCKCHAIN

The use of blockchain technology facilitates four crucial actions in the protection of the rainforest.

  1. Timber tracking

We propose a single, tamper-proof system for digital tracking timber in a supply chain that verifies the physical product. We use Blockchain technology to generate a digital ledger of transactions for trade in forest products (or potentially other commodities) that cannot be manipulated.

  1. Rainforest life class tokenization

We consider each class of life (RDFLC) as a non-fungible crypto asset. Our idea is to model this innovation through a smart contract implementation standard provided in the Ethereum specifications: ERC-721.

ERC-721 is a standard that describes the creation of non-fungible or unique tokens on the Ethereum Blockchain (or similar platforms).

A non-fungible token is used to identify something or someone uniquely.

The atomicity of RDFLC allows us to adapt our activities and increase the level of transparency with our collaborators.

  1. Incentivising

In our approach, Blockchain technology is also used to implement the concept of incentive reward (or reward unit). Indeed, the reward units are cryptographic token, which can be accumulated in return for good actions to safeguard the rainforest. Reward units can be exchanged, depending on the holder’s type, in microfinance credit, training scholarship, rainforest protector certificate, etc.

Reward units are customizable ERC-20 tokens.

Internet of Things (IoT)

To success in our activities, we manufacture some innovative and intelligent devices such as drones, smart cameras, IoT edge devices with sensors (for animals and environmental conditions, for high precision geo-location), HD map (RDFLC cartography). We also manufacture innovative blockchain connected laser-based lumber tagging stapler for timber tracking

Artificial Intelligence (AI)

We manufacture programmable IoT edge devices to capture images and monitor carbon dioxide concentration, temperature, water, humidity, luminosity, air quality, and electrochemical sensors. We provide IA edge solutions using deep learning models to make the devices self-controlled (or semi-controlled).

Main Function:

  1. OTA update firmware (or/and SW Update firmware)
  2. Waterproof case
  3. Dock station charger.
  4. Micro sim card or E sim card
  5. G sensor
  6. 4G LTE (support 5 G), BLE, Wi-Fi, RFID, 8power Harvesting Technology, quad-band: 850 /900/1800 /1900
  7. AMQP (Advanced Message Queue Protocol)
  8. Real-time tracking by SMS/GPRS
  9. Low battery alarm
  10. Temperature sensor
  11. Moisture sensor
  12. Light sensor
  13. Water level sensor
  14. Chemical sensor
  15. CO2 sensor

In addition to what cited above, we include:

  • 6th to 10th generation Intel Core processor with Intel® Iris® Pro graphics and Intel HD Graphics
  • Intel Xeon processor with Intel Iris Pro graphics and Intel HD Graphics
  • 11th generation Intel Core processor with Xe architecture
  • Intel® Iris® Xe MAX Graphics
  • Ubuntu 18.04.2 LTS (64 bit)

Data collected from sensors are used as the inputs for the embedded AI-edge solutions. The embedded deep learning models’ outputs are sent or not through cloud IoT Hub services to our PaaS for other functions.

The smart camera contains embedded IA edge solutions running computer vision deep learning models.

Main Function:

  1. OTA update firmware (or/and SW Update firmware)
  2. Waterproof case
  3. Dock station charger.
  4. Micro sim card or E sim card
  5. G sensor
  6. 4G LTE (support 5 G), BLE, Wi-Fi, RFID, 8power Harvesting Technology, quad-band: 850 /900/1800 /1900
  7. AMQP (Advanced Message Queue Protocol)
  8. Real-time tracking by SMS/GPRS
  9. Low battery alarm
  10. Temperature sensor
  11. Moisture sensor
  12. Light sensor
  13. Water level sensor
  14. Chemical sensor
  15. CO2 sensor
  16. Motion sensor
  17. Image sensor
  18. Infrared sensor
  19. Optical sensor
  20. Intel® Vision Accelerator Design
  21. FPGA (Intel® Vision Accelerator Design with Intel® Arria 10 FPGA)
  22. Ubuntu 18.04.2 LTS (64 bit)

Motion sensors detect irregular motion and alert our ecosystem to take appropriate action immediately.

HD image sensors capture high-definition images to help track human activities and animals’ movements.

Infrared sensors supplement the CO2 sensors to measure carbon dioxide levels.

Data collected from sensors are used as inputs for the embedded computer vision models and AI edge solutions. The embedded deep learning models’ outputs are sent or not through cloud IoT Hub service to our PaaS for other functions.

All the devices implement two concepts: Perception and Prediction.

Perception tasks are classification, detection, and segmentation. Applying CNN models (deep learning), the perception module allows an edge device to analyze the environment utilizing data collected through the sensors.

The Prediction module studies and predicts the behaviour of the environment and the changes possible. Prediction receives data from sensors and the perception module and then

predicts the environment changing possibilities with probabilities for those cases. The prediction module includes several deep learning models based on Convolutional Neural Network (CNN), Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM).

RainForestBrain (RFBrain)TM is a consensus mechanism (protocol), allowing the IoT edge devices to collaborate to define global perception and the global prediction for each RDFLC.

Our services

We perform specialized activities and offer customized services according to the RDFLC category.

➢ RDFLC1:

o Forest regeneration policies disclosure

o Activities zone perimeter definition

o Timber Tracking

o Monitoring of forest regeneration activities

o Certification or Labelling

o Animals’ biodiversity tracking: we must ensure that all forest species have their habitats protected.

o …

➢ RDFLC2:

o Sensibilization

o Alternative activity to generate revenue for villagers

o Monitoring villagers’ activities on the rainforest through IoT and Deep learning (Blockchain for transparency and traceability)

o ….

➢ RDFLC3:

o Monitoring of forest density

o Animals tracking

What will the donors pay for?

The donors pay for protecting or regenerate a whole live class (ownership of rainforest life class sponsorship) or a piece of a live class (shared ownership). They buy the ownership of a sponsorship. The transaction will be to buy an amount of a non-fungible crypto asset (Blockchain) representing a specific life class area (or a piece of the area). In addition to the certificate, the donors can track and verify all the activities related to their donations.