Browsing by Author "Dr. Wajahat Mahmood Qazi"

Now showing 1 - 10 of 10

A Machine Learning Technique for Motion Planning in Articulated Robots
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Moman Ali Haider; SP20-RCS-024; LHR TP 8670; Dr. Wajahat Mahmood Qazi
The integration of human-like motion into robotic systems has emerged as a pivotal research frontier, driven by the aspiration to create robots capable of nuanced interactions in real-world scenarios. This thesis delves into the development and implementation of a novel solution, leveraging state-of-the-art motion planning algorithms and neural network based technologies to instill human-like qualities in robotic motion. The research unfolds against the backdrop of contemporary challenges in the field of robotics, particularly the need for motion planning algorithms (MPAs) that not only navigate collision-free paths but also emulate human-like movements. In the landscape of motion planning, traditional algorithms like rapidly exploring random trees (RRT) and its variants have proven effective, yet their computational complexity becomes a bottleneck in higher-dimensional problem spaces. To address this limitation, the motion planning networks (MPNet) paradigm is introduced, utilizing a neural network approach with point cloud representations to navigate the intricacies of higher-dimensional environments. The computational efficiency of MPNet is harnessed to overcome challenges associated with dimensionality, offering a promising avenue for generating collision-free paths. The central problem addressed by this research is the lack of human-likeness in the paths generated by MPAs, limiting their applicability in tasks that demand human-like motion. Prior attempts to imbue human-likeness often relied on datasets recorded from human movements, leading to unreliable and constrained solutions. In response, the proposed solution adopts a hybrid approach, combining the strengths of MPNet and Artificial x VFRRT. The latter is chosen for its ability to generate human-like paths, albeit with limitations in higher-dimensional problems. The research methodology unfolds in distinct phases, beginning with the development of a reliable human-like motion dataset. The Data Generator (DG) module orchestrates this process, employing an Artificial VFRRT-based motion planner within the Kautham simulation tool. The dataset, characterized by dynamic path generation strategies and environmental diversity enhancements, forms the foundational building block for subsequent modules. The Data Encoder (DE) module steps in to transform raw obstacle representations into a point cloud format compatible with MPNet training. This adaptive encoding ensures usability and efficiency, setting the stage for the Human-like MPNet (HLMPNet) module. HLMPNet marks a paradigm shift in motion planning architectures, dynamically adapting learning parameters through an iterative process informed by human-likeness evaluations. This module is designed not only to replicate human motions but to refine and adapt its behavior based on nuanced feedback. The Human-Likeness Evaluator module acts as the discerning judge in the evaluation framework, quantifying the authenticity of generated paths. Its role in continuous learning and optimization ensures that HLMPNet evolves towards increasingly authentic and nuanced human-like motion planning. The significance of this research lies in its practical applications across various domains, including real-world human-robot collaboration, user-friendly interfaces, efficient and safe robotic operations, enhanced experiences in entertainment and services, and improved assistive and rehabilitation technologies. The proposed solution offers a holistic approach to addressing the challenges of human-like motion in robotics, contributing to the ongoing evolution of robotic systems in diverse applications
Auto World Modelling in Kautham for Motion Planning Problems
(Library Information Services, COMSATS University Islamabad, Lahore Campus., 2020-11-20) Syed Muhammad Sohail Shah Bukhari; SP16-BCS-069; Dr. Wajahat Mahmood Qazi; LHR TP 6311
Auto World Modelling in Kautham for Motion Planning Problems provides path finding solution in a dynamic environment using 3D models and Real-Time object detection with the help of Artificial Intelligence. The 3D models are generated for surface reconstruction, segmentation and model fitting. This mechanism is integrated with Real-Time object detection models which applies neural network to the acquired data. The data is then collected from hardware sensors including stereo and time of flight cameras. The process is supplemented by key points extracted from the object detections. Thus, providing better visualization and Automatic Recognition for path finding and model fitting. Models and Spatial data are then fed to the Linux based motion planning tool known as Kautham integrated with Linux framework, Robotics Operating System (ROS). ROS is used for Motion Planning, Navigation and Interaction with Robots. The Algorithms generated Kautham are then fed to ROS, enabling Robots to navigate using the most optimal path. The Union of Object detection Models (Yolo | Res-Net50) and 3D auto generated world scenes with ImageAI (python library) enables further room for precision and geometric complexity to be used in optimized planning for Kautham. This Project has limitless potential in Automating World Model construction with motion planning for wide variety of Robots currently working in fields and future deployments for Real-Time Scenarios.
Hybrid Approach to Document Attribute Classification
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Babar Imram; FA22-RCS-006; Dr. Wajahat Mahmood Qazi; LHR TP 9486
Document processing and understanding have entered a new era with the development of computer vision and its combination with natural language processing. Businesses worldwide operate on various scales, including large, medium, and small, and in various modes, including business-to-business, business-to-consumer, and business-to government. Both the seller and the buyer keep account of their transactions in business records for financial processing, inventory control, auditing, and analyzing sales trends, among other purposes. Out of all the different kinds of documents, invoices are important in the business sector. The process of obtaining and managing the information in these documents by traditional methods that rely on human labor is very costly and labor intensive. In this research work we proposed a hybrid model that utilizes deep learning for structured data extraction from invoice images. The proposed model combines YOLOv5 deep neural network which is used to detect and generate bounding boxes for the key fields in the invoice images along with Optical Character Recognition (OCR) technology which is used to extract details from those specific regions. The dataset we used in this research is a custom collection of invoices which consist of 2225 images sourced from a Danish vendor, provided by Expert System Solution for training and evaluation. Annotation was done through Makesense.ai, an annotation tool powered by AI. The proposed model was extensively validated over this custom dataset and produced excellent results with an overall accuracy of 95.8%, F1 score of 95%, precision of 100%, and recall of 98%. These results highlight the effectiveness of the proposed hybrid approach in accurately identifying and extracting textual data from invoice images. The proposed system can significantly streamline data entry processes and improve automation in invoice processing. In future work, the model may be extended to accommodate more complex invoice layouts such as tables, multilingual content, and handwritten text. Also increasing the size of data with invoices coming x from various vendors and sectors would strengthen the model, and would help generalizing its use in more close-to-reality contexts
I-Know: Knowledge Representation and Manipulation for Self-aware Robots
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Nageen Naeem; SP16-RCS-025; LHR TP 5619; Dr. Wajahat Mahmood Qazi
Knowledge representation is one of the crucial tasks while designing an artificially intelligent/self-aware agents (softbots or robots). Construction of knowledge-base and its use has been seen in past practices but with the advancement and involvement of a new generation of robotics where robot collaborate with human or with other robots. There is a requirement of knowledge-base which are not hand-crafted and the knowledge is acquired from the sensory modality interpret it and store the knowledge. The advancement in artificial intelligence (AI) design system requires explicit knowledge with the implicit knowledge. In order to extract structured knowledge from unstructured information, different information extraction techniques have been introduced. Moreover, the artificially intelligent systems are base on different cognitive architectures. These architectures at the ground level follow the same basic philosophies of cognition and regardless of school of thought the knowledge is represented using the state-of-the-art representation scheme. The utilization of these schemes are modified according to the requirement and philosophical structure of the cognitive architecture. This difference blew up the limitation of compatibility if one wants to integrate the representation scheme with other such as if representation scheme of CLARION is needed to be extended with the design rationale of atomspace they are incompatible with each other. In order to resolve this, issue this study proposed a framework for unified representation scheme which takes the best properties of the existing architecture and represents them into the atom of knowledge. The proposed representation scheme is designed to make it compatible with both softbots and robots
Multi-Context Driven Response Generation in Socio-Cognitive Agents based on Access Consciousness
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Mahmood Hussain; SP16-RCS-014; LHR TP 6396; Dr. Wajahat Mahmood Qazi
Evolution of socio-cognitive agents require them to be the co-partners with humans. Following Theory of mind (ToM) to establish and evolve this partnership an agent needs to identify humans’ mental states that may cause an agent to respond adaptively. During a social interaction, humans express their mental states that either follow the phenomena of phenomenal or access consciousness using verbal and non verbal communication. The mental states following the phenomena of Access consciousness leads to the reasoning and directing rational control upon verbal response during social interaction. Comprehension of verbally expressed information depends on understanding of its Context. Context-aware systems exist that understand the meanings behind a dialogue or conversation. However, these systems lack the ability to generate responses based on multi-contextual conversation. This study intends to design a model that could enable agents to interact with humans using verbal communication. Therefore, processing of conversational data could extract the multi context-based meanings to generate response in a naturalistic manner. Hence, the proposed model would be helpful to achieve socio-communicational capabilities based on access-conscious mental-states
Policy Based Employee Productivity Management
(Library Information Services, CUI Lahore, 2022) Rao Sharjeel; SP19-BCS-082; Dr. Wajahat Mahmood Qazi
Write Whether in the offices or factories proper monitoring of employees and workers is very important that is why supervisors are hired to make sure employees are safe and productive. But the problem with this approach is supervisors are also human beings, they can themselves make mistakes and can be irresponsible, so many people are hired just to look at other people and monitor their activities and prevent any unfortunate event from happening, but again these humans are prone to mistakes like they can fall in the trap of favouritism, jealousy, personal grudges, bribery etc. Our Policy Based Employee Monitoring System is a solution that is fully automated and is infallible to all above mentioned human problems as it gives all the control in the hand of one administrator. This means that the admin is fully responsible for all monitoring of all the employees. This Policy Based Employee Monitoring System is a system made with the mindset that it should be generic enough to be able to be deployed / installed in any industry with minimum alterations and additions. The version we are currently working on will have an interface to add policies and register users and will be able mark attendance, detect activities through cameras, record and generate alerts in case of violations (according to the policy entered) and generate daily and monthly reports of every registered employee
Programming Industrial Cobots using Natural User Interaction
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Faizan Haider , Adeel Arshad , Abdul Muazzam; SP17-BSE-108 , SP17-BSE-132 , SP17-BSE-033; Dr. Wajahat Mahmood Qazi; LHR TP 7142
Cobots are the key element of the industrial revolution. They collaborate with humans to perform repetitive tasks. Currently, cobots are being trained using Teach pendant and Lead through methods. These methods are very hectic, difficult, and costly, highly skilled labor is also required to perform these methods. This project aims to minimize these difficulties by providing a rapid programming model, that consists of a platform to train cobots using natural user interactions. The project will involve a Kinect sensor to obtain user interactions (hand gestures and verbal commands). The system will propose a solution using the inputs obtained from the Kinect sensor. This proposed solution is hybrid in which the UR robot (cobot) is simulated in WEBOTS/UR Virtual Machine that is controlled by a human in the real-world through natural user interactions. The use of a simulated environment is because of the non-availability of the physical UR Cobot.
Real-time Machine Vision-based Quality Inspection System for Small and Medium Enterprise
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Abdul Wajid , Umer Imtiaz , M. Abdul Muhib; SP17-BCS-079 , SP17-BCS-067 , SP17-BCS-103; Dr. Wajahat Mahmood Qazi; LHR TP 7157
This project aims to develop a real-time machine vision-based quality inspection system for small and medium enterprises. Labelling errors that occur during packaging can damage consumers’ perception of quality. To meet the quality and aesthetic expectations of customers, it is important to identify labelling defects before goods reach the market. All industries are in the race of making their products more apparently and qualitatively appealing. In Pakistan, Quality Inspection responsibility is handed over to the human labour but the Production lines works on so high speed that’s it is very difficult to catch minor errors in packaging by Humans so our system shall perform Machine Vision-based measures for quality assessment of the products in real-time to ensure the standard of products. Tests include identification of dirt spots, print template analysis, correct seals, torn, damaged and missing labels of packaging. The basic aim behind this project is to promote the fourth industrial revolution and Artificial Intelligence in Pakistan because our country is lacking in both. Deployment of this system will include specialized cameras, a computer with the design and implementation of industry-standard HMI.
SE42_Activity Recognition In Industry 4.0
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Muhammad Kashif , Muhammad Zubair , Anas Abdullah; FA17-BSE-155 , FA17-BSE-096 , FA17-BSE-123; Dr. Wajahat Mahmood Qazi; LHR TP 7036
Industry 4.0 is known as the fourth industrial revolution in which advanced technology like the internet of things, data analytics are used to make the industry more efficient. Industry 4.0 refers to such industrial units that are integrated and find the hidden pattern from the industry data to make it more reliable and reduce the possibility of error. With the quick growth in digital technologies, most development and research domains are entailed in the human activity recognition system. The activity recognition system will help the industry to automate the monitoring system and help the industry to analyse the performance of the employees. The activity recognition system can also allow the industry to find the hidden patterns from the activities of employees. These hidden patterns will help the industry to train its employees well and prevent their employees to do mistakes. The proposed system will supervise and monitor human activities in production and inspection processes in the Industries I-e Textile Industry. Like if a person is inspecting a product during the inspection system will check whether the person is following standards of inspection or not. The system is designed in a way that the managers or officials will be able to watch the results of activities recognized by the system on their smart devices. There are 4 actors Admin, supervisor and employee who will interact with the system. The system will be work in such a way that a Jetson device will be integrated with already installed security CCTV cameras in such a way that Jetson Units will continuously recognize human activities and sent results to the Server. The server will decide the action, and then send alerts to Supervisor, and admin. The trained model is deployed in Jetson Unit. The managerial (Admin) officials will be notified automatically if any violation is recorded by the system. Admin/supervisor will be able to take action on misconduct.
Smart Waste Management System for Clean Pakistan
(Library Information Services, COMSATS University Islamabad, Lahore Campus., 2020-11-20) Muhammad Hasnain Raza; FA16- BCS-015; Dr. Wajahat Mahmood Qazi; LHR TP 6201
This proposal presents IoT development venture of the waste management system which consists of smart bin with a monitoring system with ongoing observing framework which incorporates various technologies, for example wireless communication and other technologies. It also proposes the optimal/shortest path for garbage collectors. Point of this task is to give a proficient and cost effective waste management system the executive’s framework consequently giving perfect, sound and green condition. This examination proposed another system that empowers remote checking of strong burn through receptacle progressively by means of Wi-Fi association, to help the waste administration action.