The R&D team at Mankind Digital Solutions Company is composed of highly skilled professionals, including engineers, data scientists, and designers, who are passionate about pushing the boundaries of what is possible in the digital realm. Their collective expertise and creativity drive the development of cutting-edge solutions that address the evolving needs of businesses and individuals.

Mankind Digital Solutions Company's R&D department is committed to exploring emerging technologies, such as artificial intelligence, machine learning, blockchain, and Internet of Things (IoT). By closely monitoring industry trends and conducting in-depth research, they ensure that the company remains at the forefront of technological advancements.

Through rigorous experimentation, prototyping, and iterative development processes, the R&D team at Mankind Digital Solutions Company transforms conceptual ideas into tangible products and services. Their goal is to create digital solutions that are user-centric, scalable, and capable of driving meaningful impact in various industries.

AI/ML

AI/ML (Artificial Intelligence/Machine Learning) R&D is dedicated to advancing the capabilities and applications of AI and ML technologies. This field encompasses research and development activities that focus on improving algorithms, models, and frameworks used in machine learning and artificial intelligence systems. Scientists and engineers explore cutting-edge techniques, such as deep learning, reinforcement learning, and natural language processing, to enhance AI's ability to understand and make intelligent decisions from vast amounts of data. The outcomes of AI/ML R&D drive innovations in various domains, including healthcare, finance, robotics, and autonomous systems, revolutionizing how we solve complex problems, optimize processes, and make informed decisions in a data-driven world

  • 01 Online Classification of Skin Disease using Transfer-learning of Deep Neural Nets

    This project presents an online classification system for skin diseases using transfer learning of deep neural networks. The system allows dermatologists to upload images of skin lesions, which are then classified by a deep neural network trained on a large dataset. The proposed approach achieves high accuracy in skin disease classification tasks by leveraging learned features from general images. The system provides valuable insights to assist dermatologists in diagnosis and decision-making. With continuous learning and improvement, the online platform has the potential to enhance skin disease diagnosis, improve patient outcomes, and optimize healthcare resources.PDF

  • The speech-to-text auto-correction AI model presented in this project aims to enhance the accuracy of speech recognition systems by incorporating automatic error correction capabilities. Through the use of a deep learning architecture and a large labelled dataset, the model is trained to correct errors in speech recognition output. Techniques such as data augmentation, regularization, and fine-tuning are employed to improve performance. Evaluation using test datasets demonstrates significant improvements in accuracy and error reduction. The developed model can be integrated into existing systems or used independently, offering improved usability and communication across various domains. Future research directions include exploring multi-modal approaches and continual learning techniques.PDF

  • This project presents the implementation of a deep neural network-based fraud trademark detection system. Using a comprehensive dataset of genuine and fraudulent trademark images, a convolutional neural network (CNN) is trained to identify fraudulent trademarks. The system demonstrates high accuracy in detecting trademark infringement, providing a valuable tool for trademark authorities and brand owners. By leveraging the power of deep learning, the system effectively captures intricate patterns and variations indicative of fraud. The implementation streamlines the detection process, enhances efficiency, and contributes to the protection of intellectual property rights. Future enhancements may focus on incorporating additional data sources and refining the model to adapt to evolving fraud techniques.PDF

Electromagnetics

Electromagnetic R&D focuses on advancing our understanding and utilization of electromagnetic phenomena. It involves research and development activities aimed at improving electromagnetic technologies, such as wireless communications, radar systems, sensors, and electromagnetic compatibility. Through rigorous experimentation, modeling, and simulation, scientists and engineers in this field strive to enhance the performance, efficiency, and reliability of electromagnetic systems. They explore novel materials, antenna designs, signal processing techniques, and electromagnetic wave propagation models to push the boundaries of current capabilities. The outcomes of electromagnetic R&D drive innovations in various industries, including telecommunications, aerospace, defense, healthcare, and transportation, enabling transformative advancements in connectivity, sensing, and data transmission.

  • 01 A Dual Polarized Millimetre Wave Antenna for 5G Wireless Communication Systems Using Genetic Algorithm

    The innovative design and meticulous optimization of a dual-polarization millimetre-wave antenna tailored specifically for wireless communicationnetworks. The transmitter antenna operates with horizontal polarization, while the receiver antenna functions with vertical polarization, both effectively sharing a common electrical ground. To attain the desired Voltage Standing Wave Ratio (VSWR), a sophisticated genetic algorithm is employed to fine-tune the antenna design performance parameter using rank-based selection method. Once the VSWR requirement is successfully met, comprehensive investigations are conducted to evaluate the antenna's gain and radiation patterns. The resulting antenna design showcases exceptional performance enhancements within the frequency range of 26 GHz to 28 GHz for the transmitter and 21.50 GHz to 22.50 GHz for the receiver.PDF

  • Vivaldi antenna specifically designed for defence applications, utilizing a Genetic algorithm. The proposed antenna operates effectively within the frequency range of 380 MHz to 520 MHz and is constructed using FR4 dielectric medium. With strict limitations on physical dimensions, the antenna is confined to a size of 160 mm x 160 mm x 1 mm. To achieve the desired Voltage Standing Wave Ratio (VSWR), a Genetic algorithm is employed, employing a random search method to optimize the performance parameters of the antenna design. Once the VSWR requirement of less than 3 is successfully met, a detailed parametric analysis will be carried out to evaluate the antenna's gain and radiation patterns. The resulting antenna design demonstrates remarkable performance improvements within the designated frequency range of 380 MHz to 520 MHz.PDF

Mechanical Systems

  • 01EFFECT OF CYLINDER BLOCK FIN GEOMETRY AND MATERIAL ON HEAT TRANSFER RATE OF AIR-COOLED 4-STROKE SI-ENGINE

    The objective of this project is to increase the heat transfer rate of a 4-stroke air-cooled Spark ignition engine in order to reduce the losses (friction losses, uneven expansion of piston and cylinder, burning of lubrication oil) associated with heat produced during combustion stroke in the engine cylinder. The heat transfer rate of fins is mainly depending on three parameters, one is fins geometry and surface area of the fin, and the material used to manufacture the fins. In order to increase engine efficiency, different fin geometries like circular, rectangular wavy, and curved shapes are designed and different types of fin materials like aluminum alloy and copper alloy are used to increase the heat flux and to increase the efficiency of the engine. In this project numerical simulation of the engine cylinder with different fin geometries and fin materials is going to present to improve the heat dissipation rate of an engine. The modeling of fins is carried out by using SOLID WORKS whereas ANSYS will be used to conduct steady-state thermal analysis. The optimized design of fin geometry and fin material for the engine cylinder will be stated based on the results obtained from the analysis.PDF

  • The piston is responsible for transmitting gas pressure exerted by combustion gases into reciprocating motion and is responsible for kinetic energy. During this process, the piston crown is continuously subjected to various thermal and structural shocks in the power stroke such as gas pressure load, side thrust, and connecting rod load, inertia load. This will result in piston seizing by overheating, and cracks on top land ultimately affecting engine efficiency. To reduce heat absorption and increase insulating property Thermal barrier coating is applied on top land. The modeling of the piston is made on Solid Works. In this project, the Analysis is performed on a piston made of aluminum alloy with and without thermal barrier coating materials (Magnesium zirconate, Lanthanum doped zirconate, Yttria stabilized zirconia). Structural analysis, Steady-state thermal analysis is carried out on the piston in ANSYS and aim to study its behavior by comparing it with the conventional piston.PDF

  • In this project, we have planned to develop a wheel rim with spokes for heavy-duty vehicles similar to the present technology used for lightweight vehicles and also for two-wheelers. As we planned to develop this, we will take the reference measurements from any existing model and we develop a cad model in solid works for the existing one and also for the developed wheel with spokes. The next proceeding step in this project is the selection of material, other than the existing material like cast steel a new material like aluminum alloy or a composite is selected in which similar or better properties are extracted. In the next step of this project finite element analysis is carried out using ANSYS for static structural is followed in which total deformation, equivalent strain, equivalent stress, fatigue failure under rotation conditions, and coming to explicit dynamic analysis hitting a bump is evolved in this and the velocity, acceleration, and stress is going to be done. PDF