V. G. Pratheep

This chapter investigates the integration aspects between the Internet of Things and deep learning technologies in efforts directed toward advancing food quality monitoring, thus enhancing issues of safety and food freshness in the supply chain. IoT sensors capture real-time data with regard to environmental conditions including temperature, humidity, and gas composition all through the food production and supply process. Such data is analyzed by deep learning algorithms to detect any kind of anomalies and predict the potential hazards to enable proactive actions toward timely intervention. The chapter deals some IoT devices used for monitoring, namely smart sensors and wearable technologies, while comparing the application of deep learning models for predictive and pattern recognition analytics. Case studies underscore this integrated approach to reducing spoilage, increasing shelf life, and meeting the requirements put forth by today's safety standards.

The integration of fuel cell technology with artificial intelligence and machine learning in electric vehicles (EVs) has the potential to enhance efficiency, performance, and reliability. Fuel cells, a clean alternative to traditional engines, produce electricity through electrochemical reactions between hydrogen and oxygen, with water vapor as the only byproduct. AI-driven algorithms analyze vast data from sensors and onboard systems, while ML algorithms enable early detection of potential system failures. AI-based driver assist systems can optimize driving behaviors using fuel cell data. However, integrating fuel cell technology with AI and ML faces challenges like data management, algorithm development, and interoperability with existing vehicle systems. Interdisciplinary collaboration between automotive engineers, data scientists, and AI specialists is needed to develop robust, scalable solutions.

Robotics has been a mainstay of modern manufacturing, and it is now being employed more and more for surveillance, military, and entertainment applications. While rescue robots are employed to save humans in mining accidents, urban disasters, hostage situations, and explosions, remote-controlled surveillance robots acquire photos and video for specialised objectives. Advanced materials and effective transmissions are used in power management and new generation drive-train systems to boost speed, precision, and longevity. Both barriers with a known shape and those with an unknown shape must be overcome by the system. This requires more power and presents a design difficulty. In this chapter, the internet of things (IoT) and machine learning (ML)-integrated smart robot principles have been explained.

The robotic manipulators are nowadays used for many applications in the industries. This project involves the design and analysis of a six DOF manipulator for welding, pick and place application. We developed a robot in Solid Works and analysed it motion, load withstanding capacity and path traceability. However, design and analysis of a robot involves modelling of it forward and inverse kinematics. We modelled the forward and inverse kinematics by D-H parameters. The proposed model makes it possible to control the manipulator to achieve any reachable position and orientation in an unstructured environment. The inverse kinematics provided many possible combinations of angles for a single end effector position. A GUI was created in MATLAB for studying the forward and inverse kinematics of the robot. It gave results with precision of 0.2 cm. the load analysis also gave the maximum load it can withstand 200 KN without permanent deformation. The approach presented in this work can also be applicable to solve the kinematics problem of other similar kinds of robot manipulators.

Technology has blended itself with the everyday life of humans. People aren’t finding greater efficiency despite the field of science and technology has progressed well. Traditionally goods are moved and distributed in and around an industry manually by humans. By using an automated material handling system this problem was reduced to a certain percentage. Even though there is no efficient method to deliver the products directly to the end-user like the patients in the hospitals. So,a robotic system that transports materials all over the workplace has been designed. The serviceability of the system is based on themaster-slave concept. This system can be used in mining operations to carry materials on a rail.To make the design economical, low-cost parts are used which makes it ideal for production on an industrial scale which will aid in many applications like drug delivery systems in hospitals and also material carrying robots in mine.

This paper portrays an accelerometer as a motion sensor used to measure acceleration, vibration, and tilt angles which plays a vital role in the constructional field, inertial measurements, automotive safety system, biomedical, and yet other safety applications. A small packaged accelerometer can be made larger in size by acquiring an external power source or battery. The designed micro-accelerometer is fabricated using clean room technology, which is not quite cost effective, and even the silicon base is replaced by steel because of steel’s high yield strength and low cost to fabricate a mesoscale accelerometer. Design parameters are optimized by a Decision Tree Algorithm where the optimized dimensions are used to model the mesoscale accelerometer in INTELLISUITE 8.6 software. A Wire-cut discharger is used for the fabrication of mechanical elements of an accelerometer with a spring steel foil to reduce fabrication cost. Finally, the fabricated accelerometer with a mesoscale was made to undergo experimentation with a hammer drill and compared with simulation results to evaluate its performance in a harsh construction environment. 0.576 (μm/g) of mechanical sensitivity and 0.629 (pF/g) of electrical sensitivity is achieved.

Robust stabilization is an important characterization to get improved in the control system to optimize the performance of the desired system. Most of the controllers available suffer from problems such as difficulty in the tuning process, sluggishness in response time, quick and global convergence, etc. This paper considered proportional-integral optimized with genetic algorithm (GA-PID)controller on inverted pendulum for the control of the angle position. A MATLAB script for a GA was developed to obtain optimum PID parameters that would keep the pendulum angle at equilibrium (i.e., returns the pendulum to the desired point as quickly as possible) by minimizing an objective function (integral time absolute error [ITAE]). Furthermore, obtained outcomes confirm the performance of holding on the upright position from the downward direction such that the angle of the pendulum is manipulated to control the inverted position in its erect position in a rapid manner. In the experimental analysis, the developed GA-based PID controller is compared with the conventional PID controller to evaluate the performance of the inverted pendulum. The obtained results show that the GA-based PID controller confirms the enhanced performance indexes by holding minimum settling time and peak overshoot on comparing with the conventional PID controller.