Multirobotic systems play a significant role in several applications and agricultural field monitoring is an important area which is considered in this research work. In this paper some of the task allocation methods between robots operating in a multi-robot system is presented. The review will be helpful in understanding and development of a semiautonomous robot framework to be used in agricultural applications to reduce the burdens and make management of the agricultural field easier for farm workers. Also, the initial integration of these reviewed techniques for the semiautonomous agricultural multi-robot framework is also discussed. The next step of the research is simulation using software tools such as the Robot Operating System (ROS) 2 and Gazebo and using physical robot such as the Turtle bot to demonstrate the working of the semi-autonomous agricultural robotic framework. The research work has scope to help tasks allocations and distributions more efficient and easier to distribute among the several robots that are operating simultaneously in an agricultural field monitoring the crops and gathering field information. Also, the robot operator can use human robot interaction to control or access the robots individually in the event of safety making the multirobotic agricultural system more reliable and safer. This is a research work in progress.

In this paper, a review of some of the human robot interaction techniques used in multi-robot systems is presented. The knowledge gained from this review will help in understanding and development of an interface for human-robot interaction for the multi-robotic framework for field or environment monitoring. This is research in progress.

In this article, an unsupervised IDS (Intrusion Detection System) is presented for the detection of zero-day DDoS (Distributed Denial of Service) attacks for IoT (Internet of Things) networks that can detect anomalies without the need for prior knowledge or training in attack information. Attackers exploit existing undiscovered vulnerabilities in the system to launch zero-day attacks. There exist many traditional deep learning and machine learning based attack detection systems that cannot deal with and detect new zero-day attacks and mostly misclassify those attacks. Zero-day attacks are often new unknown threats that have not been encountered before, in addition, labelling of data is a time-consuming task for security experts, So there exists a need for unsupervised methods that can detect unseen cyber-attacks on the zero-day. DDoS attacks have recently adversely affected many organisations in terms of finance and services, as these attacks have become more sophisticated and damaging. The growth of IoT networks has facilitated the launch of more DDoS attacks. In this work, an unsupervised approach-based algorithm is proposed for the detection of zero-day DDoS attacks in IoT networks by exploiting random projection for the feature selection process to reduce the dimensionality of the network data and ensemble model consisting of K-means, GMM and one-class SVM for the unsupervised classification of the data as attack and normal using the hard voting technique. The CIC-DDoS2019 datasets are used for an extensive evaluation of the proposed method. The proposed method has obtained an accuracy of 94.55%, which is better than the other state-of-the-art unsupervised learning-based methods.

AbstractThe authors introduce an unsupervised Intrusion Detection System designed to detect zero‐day distributed denial of service (DDoS) attacks in Internet of Things (IoT) networks. This system can identify anomalies without needing prior knowledge or training on attack information. Zero‐day attacks exploit previously unknown vulnerabilities, making them hard to detect with traditional deep learning and machine learning systems that require pre‐labelled data. Labelling data is also a time‐consuming task for security experts. Therefore, unsupervised methods are necessary to detect these new threats. The authors focus on DDoS attacks, which have recently caused significant financial and service disruptions for many organisations. As IoT networks grow, these attacks become more sophisticated and harmful. The proposed approach detects zero‐day DDoS attacks by using random projection to reduce data dimensionality and an ensemble model combining K‐means, Gaussian mixture model, and one‐class SVM with a hard voting technique for classification. The method was evaluated using the CIC‐DDoS2019 dataset and achieved an accuracy of 94.55%, outperforming other state‐of‐the‐art unsupervised learning methods.

Using computer vision techniques for robotic systems made for use in agricultural settings is optimal for identifying plants yielding fruit ready to be harvested. In this study, a system for identifying objects by comparing them to an image with a single shape and measuring the total difference between them is proposed as a base for a larger system that can correctly identify the stage of maturity of tomato fruit growing on a plant based on the difference in size and shape from a base image. This was performed by a program that compares a template image with a basic shape, such as a circle or square, and counts the number of pixels yielded in a generated image that highlights the difference between the template image and another test image. Another idea proposed in this study is a program that counts objects in an image by using a technique that involves splitting an image into a celled grid. This information can be implemented into a robotic system, such as a Turtlebot, to assist in object/crop detection in an agricultural setting.

Robotic manipulators are widely used in the field of robotics. In this paper, four different manipulators are reviewed, and their use in robotic applications is presented. These robotic implementations and applications continue to evolve to meet an ever-growing industrial need for improved efficiency and functionality. This is a work in progress and the next step of this research is to use the knowledge from the papers reviewed and develop an robotic architecture using a mobile base for navigation and a manipulator for the movement of small objects in a controlled indoor environment.

Path planning and following rely deeply on mapping and object detection. In this paper, research into the application of these processes and the method to implement them are discussed and analyzed. Communication through robot operating system and MATLAB are utilized to visualize the data received and used for the programs that are run on robots. The Turtlebot serves as an educational tool with various sensors. A mapping and control method will be developed and examined to further the understanding of how each function of path planning and object detection runs and can be controlled.

Robotics is a interdisciplinary field and has seen a lot of development recently. Applications utilize multiple robots to achieve the tasks, so the concept of multi-robots is gaining popularity. Navigation and Path planning become an important topic for multi-robotic systems. In this paper, a review of some of the navigation and path planning used in multi-robot systems is presented. This is a work in progress and the next step of the research is to use this survey to develop a customized multi-robotic system for navigation in a crop field.

Robotics is a multi-disciplinary field that involves the use of multiple sensors for the application. These sensors can be of the same type or different types and measure various attributes of the environment. In this paper, a review of some of the multi sensor fusion techniques used to combine the data from the sensors for the development of robotic systems is presented.

Robotics is an emerging field and finds application in various areas. For applications such as navigation and collecting information in a large agricultural field, it is important to use more than one robot for timely and efficient information gathering. In this paper, an architecture for multi-robot navigation in an agricultural field is presented and the system uses Robot Operating System and Gazebo. This is a work in progress.

In this paper, an implementation of a human in the loop (HITL) technique for robot navigation in an indoor environment is described. The HITL technique is integrated into the reinforcement learning algorithms for mobile robot navigation. Reinforcement algorithms, specifically Q-learning and SARSA, are used combined with HITL since these algorithms are good in exploration and navigation. Turtlebot3 has been used as the robot for validating the algorithms by implementing the system using Robot Operating System and Gazebo. The robot-assisted with human feedback was found to be better in navigation task execution when compared to standard algorithms without using human in the loop. This is a work in progress and the next step of this research is exploring other reinforced learning methods and implementing them on a physical robot.

Robotic systems can play a crucial role in the agricultural field as the increasing demands for crops lead to continuous pressure for more crop quality and quantity. Agricultural work is very tedious under poor weather circumstances. The agricultural robots represent a replacement for labor in carrying out the tiresome tasks and efficiently avoiding exposing humans to health risks. The proposed work implements a ground robot to navigate the farm and monitor the plants using the Robot Operating System. The monitoring includes the classification of nine types of tomato leaf diseases and three tomato ripeness levels using Convolutional Neural Networks and computer vision using a raspberry pi camera. The model is trained on Colab, and raspberry pi3 is used to run Keras pre-trained model on TurtleBot3. Three CNN architectures are used and compared for the disease and ripeness classification of tomatoes.

Robotics is used in various applications in today's world and these systems are complex in design and working. In order to make the robotic system dependable and more reliable it is important to handle the faults and act in a timely manner. Thus, a fault tolerant and self-healing feature must be integrated with the robotic system to make it more dependable. In this paper, the continuation of the research work of such as robotic system is presented. The focus of the paper is to show the development of the research work through the implementation, some of the test scenarios and the results using Robotic Operating System (ROS). This research is a work in progress and the next steps will be the hardware implementation of the robotic system on a Turtlebot robot.

Fault tolerance is important in a robotic system architecture because of the advancements in robotics which makes the robots much smarter and more capable on executing tasks. Thus fault tolerant and self-healing robotic control systems help prevent any cascade errors that can lead to the robot becoming inoperable or causing a safety hazard. The current paper is a work in progress describing the initial efforts of implementing a fault tolerant robotic system architecture and presents initial progress of this implementation where the robotic control system demonstrates a fault tolerance and a self-healing algorithm using human robot interaction in the process.

In this paper, an overview of the use of machine learning algorithms on autonomous robotic systems is presented. The concept of hardware and software integration and development is explained through a small robot such as Turtlebot used in the examples taken. This paper presents a survey of how machine learning can be utilized on a hardware platform and is the initial stage of the research work. This is a work in progress and the next stage is using this literature survey towards developing research towards a comparison of machine learning algorithms on the Turtlebot.

In this paper, the robot is exploring and creating a map of the environment for autonomous navigation. Exploration forms an important role in creating the map and locating the obstacles for path planning. Autonomous Frontier Based Exploration is implemented on both hardware and software of the Turtlebot3 Burger platform. For Simultaneous Localization and Mapping (SLAM), the Breadth-First Search (BFS) algorithm is used. The path planning uses the A * algorithm for the global planning and for local planning uses the Dynamic Window Approach (DWA). The software used in this work is Ubuntu 18.04, along with ROS Melodic on virtual box. This is a work in progress, and the results will be used in further research work for autonomous monitoring of the agricultural fields.

With the increasing demand for crops, maintaining high efficiency with large quantities of production is required, and the traditional methods to control the farm is getting tedious and prone to decision making mistakes. Therefore, using robotics to carry on these tasks is useful, as robots can work under adverse weather conditions, and this will protect the farmworker from exposure to health risks, especially that skin cancer is more common among farmers. In this paper, a survey of the implementation and navigation of autonomous aerial and ground robots for monitoring plant health, ripeness, and soil moisture using different approaches is presented. This is a work-in-progress, and these techniques will be implemented on a low-cost affordable robotic vehicle such as the Turtlebot to perform these tasks efficiently, thereby minimizing the presence of human workers on the field during adverse weather conditions hence protecting their health and well-being.

In this paper an initial implementation of a sensor fusion based robotic system architecture for motion control is presented. The architecture also consists of a human robot interaction interface to help the robotic system during the application. This is a work in progress and the paper presents the initial efforts covering the architectural implementation of the system that includes the experimental setup and the block diagram of the architecture.

Since the robotic topic is evolving every day, and humans are in constant interaction with robots that use different sensors and different environment, it is needed to take into consideration the importance of sensor fusion for motion control. The method of sensor fusion consists of merging the data from different sensors to obtain more accurate data for object detection as well as parameters and statements required for the auto location, mapping, path planning, motion control, and motion execution. There are different ways to make possible the human-robot interaction, but no matter the concept implemented the safety of the human and the robot needs to be one of the priorities of the robotic system. In this paper the continuation of the research work is presented from the previously proposed robotic system. The robotic system is extended to include a safety controller and priority module. Also, the improved architecture will be able to receive commands from multiple human inputs, and it will determine which command will have the higher priority to be executed. The paper focuses on the continuation of the research work through detailed implementation, some test case scenarios and their results taking into consideration that the robot can execute a safety measure. The simulations are done using ROS and Gazebo.

Human-Robot Interaction is an emerging field in the area of robotics and growth in this field finds application in several areas of the society. In this paper an initial progress of the design of a robotic system for elderly care is proposed. The paper describes the basic architecture of this proposed system and presents an initial implementation of this robot. This is a work in progress, and the robotic system will be updated and developed as the research progresses when more sensors and other components are interfaced providing more functionalities to the robot. Also, another goal of the project is contribution in undergraduate electrical and computer engineering education where students understand and learn to work on both hardware and software aspects in the design of an assistive robotic system.