Please describe your proposed solution.
This proposal is a continuation of our proposals in FUND 5 and FUND 8 ( Ongoing ), in which we created a proof-of-concept piece of software to upload trained and encrypted messages/images on the Cardano blockchain. Some Universities that have some units of our robot are currently testing our company's repositories based on the Cardano blockchain.
- Oxford University
- Stanford University
- Osaka University
- Cambridge University
- YONSEI UNIVERSITY ( Seoul )
- Physics and Electronics University of York
- Seoul National University
- UPM in Madrid
- Freie Universität Berlin
- and many more….
The Problem in Context
Human decision making is often irrational and biased. Group decisions have dynamic interactions that further complicate the choice process and frequently result in outcomes that are suboptimal for both the individual and the collective, and for a trained artificial intelligence system in a centralized system it can be easily hacked.
Example to understand easily:
- In Robotics: Imagine a swarm of "Cop Robots" that patrol the streets greeting people and looking ( OpenCV VISION ) for antisocial behaviors. These robots could communicate (VOICE RECOGNITION or CLOUD COMPUTING ) over a blockchain and have action-triggers with smart-contracts. These could run when they spot one person hurting another, to have the system vote on the best strategy to approach the scene or to call for help. But to achieve this type of behavior, it is necessary that smart-contracts have improved security and are able to interact with information from outside the blockchain. Imagine if someone is hacking a centralized and private cloud service and decides to shoot.
- Valid for Autonomous Cars: Kill the pedestrian or the passenger? The complicated ethics of self-driving cars. Imagine this scenario: You’re driving down the road when suddenly, out of nowhere, a lone pedestrian jumps into your path. The road is narrow and walled in by a barrier on either side, so your options are limited. You can either keep driving straight ahead and hit the pedestrian, or you can save the pedestrian by swerving into one of the barriers. What do you choose?
Now imagine the scenario where there are five or ten pedestrians on that road instead of one. Does that answer change? What if they were kids?
In the real world, rare though such scenarios might be, the driver would likely be forced to make a split-second decision about how to behave – a choice that he or she probably wouldn’t even have time to fully think through.
But coming up with responses to these situations becomes a little more pressing when we consider the possibility that self-driving vehicles might one day be making these decisions for us – which means engineers must figure out the best and most ethical ways to program them for such situations ahead of time.
SOLUTION
In facilitating the decision process for Robotics and that autonomous vehicle in situations like these, the blockchain can be part of the answer as it can have millions of trained decisions under the AI/ML decision library and accept the best decisions - even at the expense of the passenger (well, at least the kids live to see another day and you could still have a high probability of surviving the crash). We can make this happen in private and public servers with the proof of concept of each decision uploaded to the Cardano Blockchain.
Once the new Open Source API for our SDK is finalized, we can send it to all our community so that they can validate and provide feedback for them. There will be a ready set of users for the new API/SDK that link to the Blockchain.
Please describe how your proposed solution will address the Challenge that you have submitted it in.
In creating this decision making system on the Cardano Blockchain, we are able to address two challenges:-
- Proof of concept for a decentralized decision making system on a blockchain.
- Proof of value that a decentralized decision making system on a blockchain is more trustworthy than a centralized system.
We believe our project will open doors for use and validation of such architecture on the Cardano blockchain in future.
How can we incorporate new developers to the Cardano Blockchain?
We can start with developing an Open Source API and SDK for AI/ML applications in Natural Language and Speech Recognition which can be tested with our Open Sourced robots. This forms the baseline for integration with other tool sets to test AI/ML applications and algorithms with training data on the blockchain.
Over the last 14 years, we have a wide and strong community in various universities and research centers for our open source robots. They help build Q.bo's open source platform on GITLAB. Our existing SDK allows us to develop Computer Vision, Natural Language and Speech Recognition applications.
In this third phase ( FUND9 ) of development, we want to create a new intermediate layer (API) that serves as an intercom between the operating system and the robot to take and make deccissions. This will allow all these centers that work with our robot to develop new applications on the Blockchain and have it tested on the robot.
This will be the baseline model of how blockchain can be used to validate and acquire data models for AI/ML applications.
How are we going to reach new developers in the future?
Once the new Dapp for our SDK is finalized, we can send it to all our community so that they can validate and provide feedback for them. It is a ready set of users for the new API/SDK that link to the Blockchain.
In 2017 we launched a crowdfunding campaign on Indiegogo with great success since in just over 1 month we raised more than $ 200,000.
- INDIEGOGO CAMPAIGN
Our intention is that throughout 2022 or 2023 we will launch another new campaign with the fully finished SDK with support for Cardano Blockchain in the two largest crowdfunding companies in the world, Kickstarter and Indiegogo, which will give greater visibility to the Cardano project and attract new users and developers to the ecosystem that is constantly expanding and evolving.
What are the main risks that could prevent you from delivering the project successfully and please explain how you will mitigate each risk?
- Challenge: Sending messages (large text or images) may be expensive in terms of currency. Possible Solution: Hybrid Model: Heavy information stored in a public database, certification upload to the Cardano Blockchain.
- Challenge: Synchronization between multiple nodes (robots) may be problematic. Possible Solution: Create a queue system to manage the message receiving/sending.
- Challenge: Adaptation to Raspberry Pi 4. Possible Solution: Optimize and test all the system for a Raspberry Pi.
- Challenge: Public databases are not stable over time. Possible Solution: Use DOI to always assure the access to critical items.
- Challenge: Deploy the same configuration over different units. Possible Solution: Use of Nix and .nix files.