RV Institute of Technology and

Management®

Robotics and Robotics as an Application of AI

Robotics is an engineering branch focused on the conception, design, manufacture, and operation of
robots. It is an inter-disciplinary field encompassing electronics, computer science, artificial intelligence,
mechatronics, nanotechnology, and bioengineering. The term 'robotics' was first used by science-fiction
author Isaac Asimov in the 1940s, who also proposed three fundamental rules for robots:
1. Never harm humans.
2. Always follow instructions from humans without violating rule
3. Protect themselves without violating the other rules.

Key Aspects of Robots:

• They possess electrical components for power and control of machinery, typically drawing current from
a battery.
• Their mechanical construction, shape, and design are determined by the specific task they are intended to
accomplish.
• They are programmed with instructions that dictate what, when, and how they perform tasks.

Artificially Intelligent Robots (AIRs):

• While many view AI and robotics as interchangeable, they are distinct fields with a small overlapping
area where artificially intelligent robots (AIRs) exist. AIRs serve as the bridge between robotics and AI.
• Most industrial robots, for instance, perform repetitive movements and do not require artificial
intelligence. However, to perform complex tasks, an AIR must implement AI algorithms.

• Examples of AI guiding robot functionality include:

 RV Institute of Technology and
Management®

Warehousing robots using path-finding algorithms for navigation, drones using autonomous navigation to
return when batteries are low, and self-driving cars employing AI techniques to detect and avoid potential
hazards on the road.

Types of Robots: Robots can be broadly categorized based on their capabilities and design:

1. Pre-programmed robots: These operate in controlled environments to perform simple, repetitive

tasks more efficiently than humans, such as a mechanical arm on an automotive assembly line for
welding or inserting parts.
2. Humanoid robots: Designed to resemble humans and/or mimic human behaviour, capable of
human-like activities like running, jumping, and carrying objects. Sophia and Atlas are notable
examples.
3. Autonomous robots: Operate without human operators, designed for tasks in open environments
that do not require human supervision (e.g., the Roomba vacuum cleaner that roams freely).
4. Teleoperated robots: Mechanical bots controlled remotely by humans, used in extreme conditions
where human presence is risky (e.g., human-controlled submarines for underwater pipe leaks or
drones for detecting landmines).
5. Augmenting robots: These enhance existing human capabilities or replace lost ones, such as
robotic prosthetic limbs or exoskeletons used for lifting heavy weights.
Uses of Robotics: Robots are utilized in various sectors, including:

1. Assisting in fighting forest fires and performing search-and-rescue missions after natural
disasters.
2. Working as cobots alongside humans in manufacturing plants, warehouses (for shipping,
handling goods, quality control, last-mile delivery), and other workspaces.
3. Providing companionship to the elderly.
4. Serving as surgical assistants and delivering medicines or linens in healthcare.
5. Performing household tasks like vacuuming and mowing the grass.
6. In the automotive industry, self-driving cars (e.g., Tesla, Waymo) integrate data science
with robotics to offer autonomous travel.
7. Space agencies like NASA use robots (e.g., Mars rovers Spirit and Opportunity, Cassini)
for exploration, moving large objects, and scouting new areas. Robonaut is being
developed to assist astronauts in space.
8. Performing tasks like diffusing bombs.

Drones Using AI

A drone, or 'unmanned aerial vehicle' (UAV), is an unpiloted aircraft that leverages technologies
like computer vision, artificial intelligence, and object avoidance. Drones vary greatly in size and are
increasingly deployed for dangerous and high-paying jobs.
 RV Institute of Technology and
Management®

Applications of Drones Using AI:

1. Emergency Response: Drones with thermal cameras identify victims.

2. Humanitarian Aid/Disaster Relief: Assess damage, locate victims, deliver aid, and
prevent disasters.
3. Surveillance: Monitor forest fires using thermal cameras.
4. Conservation: Geospatial imagery for monitoring and tracking wildlife to combat
poaching.
5. Disease Control: Capture and test mosquitoes, and deliver vital drugs/medical equipment
to remote areas.
6. Bomb Detection: Small drones with cameras can penetrate constricted spaces to detect
bombs.
7. Air Strikes: Used for military operations, though raising ethical concerns due to lack of
accountability.
8. Agriculture: Gather data, automate processes, improve efficiency, and predict harvests.
9. Weather Forecasting: Collect data on climatic parameters to predict global weather
changes.
10. Maritime: Inspect ships and identify emissions infractions.
11. Waste Management: Clean oceans, collect waste in ports, and maintain wastewater
systems.
12. Energy: Survey areas for new energy production sites, detect leaks with thermal sensors,
and design solar array configurations.
13. Mining: Capture large data amounts from the air for measurement and assessment,
reducing risks for surveyors.
14. Construction Planning: Improve planning, project monitoring, and site management.
15. Urban Planning: Gather data in populated areas to implement data-driven improvements
like suggesting green spaces.
16. Personal Transportation: Autonomous aerial vehicles (AAVs) for passenger ferrying,
aiming for safe landings in urban environments (e.g., EHANG, Uber, Airbus, Boeing,
Rolls-Royce's "robotaxis").
17. Space Exploration: NASA uses drone-like helicopters for Mars missions (e.g., Mars
2020) and nuclear-powered drones for exploring moons like Titan.
18. Telecommunications: Inspect towers and guide repair teams to restore service.
19. Internet: Provide Internet access in remote areas (e.g., Facebook's Aquila, SoftBank's
"floating cell towers").
20. Outdoors: Aerial photography and mapping for activities like climbing and skiing.
21. Tourism and Hospitality: Offer luxury accommodations in new or remote locations, and
deliver packages/room service.
 RV Institute of Technology and
Management®

22. Live Entertainment: Used by companies like Disney for synchronized light shows and
floating projection screens.
23. Journalism/News Coverage: Gather news from hard-to-reach areas.
24. Food Services: Facilitate faster and cheaper online food delivery.
25. The Indian government has also used drones to deliver COVID vaccines to remote areas.

No Code AI
No-code AI refers to tools that enable individuals to build applications and systems without writing
extensive code, relying instead on visual interfaces and guided user actions. These tools are often pre-
integrated with others for seamless information exchange.

Applications Built with No-Code Tools:

• Websites and landing pages (e.g., Web flow).
• Web or mobile applications (e.g., Bubble, Adalo, Mendix, Thunkable).
• Chatbots or virtual assistants (e.g., Octane AI, Kore.ai, Landbot, mindsay).
• Databases (e.g., Airtable).
• Connecting tool stacks (e.g., Zapier, tray.io, Integromat, Parabola, Paragon).
• E-commerce platforms (e.g., Shopify, Weebly).
• Membership management (e.g., Memberstack).
• Newsletters (e.g., Mailchimp, Mailjet).

Why No-Code AI is Beneficial:

• Enables automation through plug-and-play or drag-and-drop user interfaces.

• Allows users with no coding skills to utilise AI algorithms for business problems.
• Helps businesses build AI systems without significant time and resource investment.
• Increases accessibility of data science and AI for small and mid-sized companies.
• Enhances usability, allowing non-technical users to create AI solutions cost-effectively and quickly.
• Provides solutions of good quality and reduces human errors in system setup.
• Examples include Google AutoML, which facilitates training high-quality ML models tailored to
business needs.

Categorisation of No-Code AI Tools by Technology:

• Computer Vision: For machines to obtain and act on information from digital images, videos, and
PDFs.
• NLP (Natural Language Processing): For machines to understand and process spoken and written
language, such as text messages.
• Predictive Analytics: For predictive modelling based on structured (tabular) data, like predicting churn
rates or stock prices.
 RV Institute of Technology and
Management®

Future of No Code AI:

• Research suggests that by 2024, nearly 65% of application development will occur on low-code or no-
code platforms, with no-code AI playing a significant role.
• It allows non-AI experts to implement and test ideas, and AI experts to create ML solutions faster.
• Supports collaboration between AI and domain experts.
• Aims to make AI technology adoptable universally.

Reasons for Using No-Code AI:

It addresses critical problems such as:

• Gap Between AI Experts and Domain Experts: No-code platforms bridge this gap, allowing domain
experts to test ideas and communicate issues effectively with AI experts, thereby increasing the success of
AI initiatives.
• Limited AI Experts and Abundance of Problems: As AI experts are few and costly, no-code AI
platforms empower domain experts to find AI solutions independently. It's noted that while only 0.25% of
the global population can code, anyone can quickly learn to use no-code tools.
Low Code AI
Low-code AI is a software development technique introduced in 2011 to accelerate application
development with minimal coding. It supports visual development through intuitive modeling and
graphical user interfaces (GUIs).

Core Principles and Functionality:

• Based on model-driven design, automatic code generation, and visual programming.

• Provides integrated tools to eliminate the need for line-by-line code writing.
• Users can design applications by drawing flowcharts in a visual editor, with the code
being automatically produced.
• Applications can be created by dragging and dropping pre-defined components.
• Popular low-code platforms include Mendix, Outsystems, Creatio, Appian, and Creator.
• Low-code solutions promote fast application delivery, minimal effort for installation/configuration, and
reduced training/implementation time.
• They also offer built-in security and maintenance, leading to cost and time savings, and lower
development risks with high ROI.

Low-Code vs. No-Code AI Development:

 RV Institute of Technology and
Management®

• Low-code requires minimal programming, whereas no-code requires no programming at all.

• Both use building blocks, but low-code offers the flexibility to integrate custom software as building
blocks.
• Low-code tools are better suited for teams with some data expertise.
Users of Low-Code Development:
• AI Beginners: Makes AI technologies more accessible, reducing the need for full-time AI developers.
• Developers: Helps AI developers (e.g., Computer Vision Engineers) to build AI applications and
integrate custom solutions, allowing them to focus on optimizing functionality rather than repetitive
coding.
• Researchers: Facilitates rapid prototype development, speeds up experiments, and enables testing
changes without expensive infrastructure, along with easy comparison and benchmarking of approaches.

Low-Code AI Platforms for Computer Vision:

• Viso.ai is an example of a low-code AI platform specifically designed for automating computer vision
applications.
• It enables the creation of custom computer vision and deep learning applications that process real-time
video feeds using deployed AI algorithms, leveraging pre-trained AI algorithms and pre-built modules for
quick solution development and continuous optimization.

Components of Low-Code AI Platforms:

• Graphical User Interface (GUI): An intuitive interface with clear visual elements for non-technical
users to build AI vision applications.
• Pre-Built Integrations: Ready-to-use building blocks for functionalities like camera input and pre-
trained AI models, ensuring secure data storage and communication.
• Application Manager: Tools for testing, deployment, and ongoing maintenance, including code
updates, bug fixes, and performance improvements.

Disadvantages of Low-Code/No-Code Platforms:

• Security: Potential for compromised security if platforms lack adequate access protocols, requiring
thorough review of terms and conditions.
• Lack of Customization: Limited functionality due to being designed for specific problems.
• Requires Consultation or Training: Users, including ML engineers, often need training and guidance.
• Lack of Trust: Traditional ML still has more support and libraries, affecting trust in LCNC platforms.
• Lock-In Strategy: High switching costs can lead to dependence on a single vendor.
• Limitations on Personalization: Some solutions do not allow users to change certain parameters.
• Data Management: May still require expertise from data scientists and engineers.
• Scalability: Currently limited in creating scalable solutions for complex problems.
 RV Institute of Technology and
Management®

Future of Software Development with Low-Code:

• Low-code AI solutions are profoundly impacting the market and are rapidly expanding.
• Forecasts indicate that by 2024, over 65% of applications will be developed using LCNC approaches,
with large enterprises adopting at least four such tools.
• The global low-code development technologies market is projected to reach $65 billion by
2026 and $187 billion by 2030.
• It benefits data-driven sectors like marketing, sales, and finance by enabling predictions (e.g., churn
rates), report analysis, smart suggestions, and automated invoicing.
• Low-code platforms offer a better scope for customization compared to no-code tools.