Browsing by Author "Sen Gupta, Gourab"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Autonomous agents in a dynamic collaborative environment : a thesis presented in partial fulfilment of the requirements for the degree of PhD in Engineering at Massey University, Palmerston North, New Zealand
    (Massey University, 2008) Sen Gupta, Gourab
    The proliferation of robots in industry and every day human life is gaining momentum. After the initial few decades of employment of robots in the industry, especially the automotive assembly plants, robots are now entering the home and offices. From being pick-and-place manipulators, robots are slowly being transformed in shape and form to be more anthropomorphic. The wheeled robots are however here to stay for the foreseeable future until such time as artificial muscles, and efficient means to control them, are well developed. The next phase of development of robots will be for the service industry. Robots will cooperate with each other to accomplish collaborative tasks to aid human life. They will also collaborate with human beings to assist them in doing tasks such as lifting loads and moving objects. At the same time, with the advancement of hardware, robots are becoming very fast and are capable of being programmed with more intelligence. Coupled with this is the availability of sophisticated sensors with which the robots can perceive the real world around them. Combinations of these factors have created many challenging areas of research. Several factors affect the performance of robots in a dynamic collaborative environment. The research presented in this thesis has identified the major contributing factors, namely fast vision processing, behaviour programming, predictive movement and interception control, and precise motion control, that collectively have influence on the performance of robots which are engaged in a collaborative effort to accomplish a task. Several novel techniques have been proposed in this thesis to enhance the collective performance of collaborating robots. In many systems, vision is used as one of the sensory inputs for the robot’s perception of the environment. This thesis describes a new colour space and the use of discrete look-up-tables (LUT) for very fast and robust colour segmentation and real-time identification of objects in the robot’s work space. A distributed camera system and a stereo vision using a single camera are reported. Advanced filtering has been applied to the vision data for predictive identification of the position and orientation of moving robots and targets, and for anticipatory interception control. Collaborative tasks are generally complex and robots need to be capable of exhibiting sophisticated behaviours. This thesis has detailed the use of State Transition Based Control (STBC) methodology to build a hierarchy of complex behaviour. Behaviour of robots in a robot soccer game and features such as role selection and obstacle avoidance have been built using STBC. A novel methodology for advanced control of fast robots is detailed. The algorithm uses a combination of Triangular Targeting Algorithm (TTA) and Proximity Positioning Algorithm (PPA) to position a robot behind an object aligned with a target. Various forms of velocity profiling have been proposed and validated with substantial test results. The thesis ends by looking at future scenarios where robots and human beings will coexist and work together to do many collaborative tasks. Anthropomorphic robots will be more prevalent in future and teleoperation will gain momentum. Throughout the thesis, the engineering applicability of proposed algorithms and architectures have been emphasised by testing on real robots.
  • Thumbnail Image
    Item
    Non-destructive and cost-effective 3D plant growth monitoring system in outdoor conditions : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in School of Food and Advanced Technology at Massey University, Palmerston North, New Zealand
    (Massey University, 2022) Paturkar, Abhipray P.
    Plant growth monitoring is one of the crucial steps within plant phenotyping. Traditional manual measurement techniques are error-prone and destructive. In recent times there has been substantial progress in computer vision-based methods. Due to their non-destructive nature and increased accuracy, imaging techniques are becoming state-of-the-art in plant phenotyping. However, most of the associated cameras, sensors, and processors are expensive, resulting in their reduced applicability in this area. This thesis proposes a framework for low-cost plant growth monitoring. A novel, cost-effective and non-destructive 3D method is proposed. It uses a smartphone’s camera and is based on the structure-from-motion algorithm to construct 3D plant models. This algorithm uses several overlapped images to build the model. The modelling speed and quality largely depend on the number of input images used. It is challenging to select the right number of images to generate an accurate plant model - too few images might generate false points in the 3D point cloud, whereas too many images will result in redundant processing, which will inevitably increase computation time. An analytical method is proposed to determine the appropriate number of images for modelling to solve this problem. Once the 3D model is generated, it is essential to segment the various plant components such as leaves and stems to measure traits. The segmentation method needs to be able to work regardless of the particular plant architecture. This thesis proposes a segmentation method using Euclidean distance to segment the point cloud. Finally, plant traits for growth monitoring are measured: leaf length, leaf width, number of leaves, stem height, and leaf area. Methods to accurately measure leaf length, width and stem height when curled are proposed. To conclude, this thesis demonstrated that the proposed framework could monitor plant growth and calculate structure and growth parameters in different outdoor conditions. The framework was tested using five different plants with different architectures: cauliflower, lettuce, tomato, chilli, and maize. Hence, this framework is a potential alternative to costly state-of-the-art systems.