Results in Journal EAI Endorsed Transactions on AI and Robotics: 13
(searched for: journal_id:(7021748))
Published: 7 January 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1; https://doi.org/10.4108/airo.v1i1.2709
A few years back, when the image processing hardware and software were created, it was limited, and most of the time, object detection would fail., But as with time, the advancement in technology has significantly changed the scenario. A lot of researchers worked in this field to carry out a solution through which they can detect objects in any field, especially in the robotic domain . In today's world, with so much research in the field of deep learning, it is very easy to identify and detect any object using computer vision. This paper focuses on the various deep learning technologies and algorithms through which object detection can be done. A new and advanced deep learning technology known as salient object detection has been discussed. Also, the 3D object detection and the end-to-end approach for object detection are discussed. The existing methods of deep learning through which object detection can be done. The applications in which object detection is being used and the importance of object detection. It also reports; what the predecessors have done, what problems have been solved by them, how they solved these problems, the characteristics of the predecessors' methods and their future work.
Published: 7 January 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1; https://doi.org/10.4108/airo.v1i1.551
Gimbal or other stable platforms have structures that move according to its functions. This is for the purpose of keeping track of the goals to the fullest. Tracking targets can become difficult as the subject moves further and further away and they are out of the gimbal’s allowable viewing range. Besides, under the influence of noise signals form outside space, it becomes even more difficult to observe the gimbal’s targets. To overcome above disadvantages, this paper is presented an adjustment method to limit above risks. Adjusting Linear Quadratic Gaussian (LQG) for expensive gimbal systems, noise signals are processed purely by Kalman filters to improve the function of observing targets. In addition, proportional- integral-derivative (PID) controller, artificial neural network in this case is also considered to verify the effectiveness of control methods listed below. In particular, ANN is the most effective control method today to deal with unwanted signals. These unwanted signals can cause worsening conditions during the operation of systems.Therefore, artificial network (ANN) is a solution to information and communication security problems. Simulation is done by Matlab. Novelty of the work: no previous research has been published for this genre. The study of this genre with the use of artificial intelligence is suggestive of the study of artificial intellligence technologies at a higher level. This category is also a suggestion for studying a smoother control method based on existing data.
Published: 13 July 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-7; https://doi.org/10.4108/airo.v1i.1124
The constant false alarm rate (CFAR) algorithm is a strong technique to detect and track dynamic targets in an environment of an unknown noise floor. Multiple reflections of a pulse from a target and different signal processing techniques applied to the received pulse, make it spread along the range and/or Doppler axis. Spreading of a pulse results in a cluster of targets detection for a single target when the CFAR technique is applied to it. This causes difficulties in calculating those target’s parameters which require only a single maximum peak for a target, such as Radar cross-section (RCS), relative phase, etc. This manuscript proposes a solution, which extracts a single independent peak for a target that had clusters of peaks after CFAR. The novelty of the algorithm is that it works well to extract a single peak for each of all targets in the multiple targets environment, as compared to the conventional global maxima finding techniques which outputs only one target of the maximum amplitude while suppressing the rest of the small targets. The algorithm is basically a local maxima finder algorithm termed as peaks detector algorithm. An attractive feature of this algorithm is that it neither disturbs the Probability of false alarm rate (Pfa) of CFAR nor it affects the probability of detection (Pd) of a target. The algorithm is tested and its performance is evaluated in a multiple targets environment on the output of 1D and 2D CFAR.
Published: 10 June 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-10; https://doi.org/10.4108/airo.v1i.656
Trajectory tracking and obstacle avoidance lies at the heart of autonomous navigation for mobile robots. In this paper, a control architecture for trajectory tracking while avoiding obstacles and controller tuning is proposed for a differential drive mobile robot (DMR). The framework of optimization algorithm is inspired by the food search behavior of beetles using their antennae. Path planning and controller tuning remain computationally demanding tasks despite of the proposed algorithms existing today. Our bio inspired approach unifies these two problems by minimizing the respective cost functions and solving the optimization problem efficiently. Trajectory tracking problem is based on the difference of the current and next pose of the robot while obstacle avoidance is achieved on the principle of maximizing the minimum distance between the robot and obstacle in the path of the robot. The proposed architecture is simulated in V-REP environment using MATLAB. Simulation results have verified that beetle antennae search can successfully plan and track the reference path by tuning the PID controller efficiently.
Published: 18 May 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-13; https://doi.org/10.4108/airo.v1i.383
This paper explicates hybrid optimization driven Extreme Machine Learning (ELM) strategy is developed with feed forward neural network (FFNN) for the classification of data and improving ELM. The pre-processing of input data is carried for the missing value imputation and transformation of data into numerical value using exponential kernel transform. The significant feature is determined using the Jaro–Winkler distance. The classification of data is done using the FFNN classifier, which is trained with the help of the hybrid optimization algorithm, namely developed modified Cuckoo Search and Invasive Weed Optimization (CSIWO) ELM. The modified CSIWO is devised by integrating the modified Cuckoo search (CS) algorithm and Invasive Weed Optimization (IWO) algorithm. The experimental results proposed in this paper show the feasibility and effectiveness of the developed CSIWO ELM method with encouraging performance compared with other ELM methods.
Published: 18 February 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-6; https://doi.org/10.4108/airo.v1i.16
Fork/Join is a simple but effective technique for exploiting the parallelism. When developing a parallel program using Fork/Join, one of the main things is how a large task is decomposed into subtasks whose results can be combined as a final result. In this paper we show how to develop Fork/Join parallel programs through refinement and decomposition. We take Fork/Join style task decomposition as a refinement which we call Fork/Join refinement. Proof obligations of refinement can ensure the correctness of decomposition. For practical application, we provide a refinement pattern for the Fork/Join refinement and extend an atomicity decomposition diagram to illustrate it. Our approach provides a good framework for modeling Fork/Join parallel programs and showing proof obligations of correctness for such programs. We illustrate the approach by applying it on a small case.
Published: 13 January 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-8; https://doi.org/10.4108/airo.v1i.15
Optimization algorithms are commonly used in the industry. The optimization strategy, if key elements are ignored, can quickly render the solution unfeasible. As a result, various optimization strategies are applied at all aspects of the industry level. The switched reluctance motor is the most affordable of all motor types. The high torque density attribute of induction motors is one of the market's major drivers. Switched reluctance motors are also employed in high-volume and high-starting torque appliances. The Smart Bacterial Foraging Algorithm (SBFA) mimics the chemotactic behavior of E. Coli bacteria for optimization purposes. This method is used to calculate the coefficient of a typical Proportion–Integration (PI) speed controller for SRM drives while accounting for torque ripple reduction. The results of the modeling and experiments reveal that the modified PI controller with SBFA performs better. The proposed optimization strategy results in increased performance when compared to regular BFA.
Published: 23 February 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-19; https://doi.org/10.4108/airo.v1i.20
After nearly 30 years of development, service robot technology has made important achievements in the interdisciplinary aspects of machinery, information, materials, control, medicine, etc. These robot types have different shapes, and mainly in some are shaped based on application. Till today various structure are proposed which for the better analysis’s need to have the mathematical equation that can model the structure and later the behaviour of them after implementing the controlling strategy. The current paper discusses the various shape and applications of all available service robots and briefly summarizes the research progress of key points such as robot dynamics, robot types, and different dynamic models of the differential types of service robots. The current review study can be helpful as an initial node for all researchers in this topic and help them to have the better simulation and analyses. Besides the current research shows some application that can specify the service robot model over the application.
Published: 4 February 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-13; https://doi.org/10.4108/airo.v1i.17
The proportional–integral–derivative (PID) control systems, which have become a standard for technical and industrial applications, are the fundamental building blocks of classical and modern control systems. In this paper, a three-layer feed-forward neural network (NN) model trained to replicate the behavior of a PID controller is employed to stabilize control systems through a NN feedback controller. A novel bio-inspired weights-and-structure-determination (BIWASD) algorithm, which incorporates a metaheuristic optimization algorithm dubbed beetle antennae search (BAS), is used to train the NN model. More presicely, the BIWASD algorithm identifies the ideal weights and structure of the BIWASD-based NN (BIWASDNN) model utilizing a power sigmoid activation function while handling model fitting and validation. The results of three simulated trials on stabilizing feedback control systems validate and demonstrate the BIWASDNN model’s exceptional learning and prediction capabilities, while achieving similar or better performance than the corresponding PID controller. The BIWASDNN model is compared to three other high-performing NN models, and a MATLAB repository is accessible in public through GitHub to encourage and enhance this work.
Published: 2 February 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-7; https://doi.org/10.4108/airo.v1i.18
Redundant manipulators are widely utilized in numerous applications among various areas in industry and service. Redundant manipulators take advantage of their inherent or acquired redundancy to achieve certain benefits in kinematic control. Different from non-redundant manipulators, optimization paradigms are more likely to be established and may be more efficient for kinematic control issues in redundant manipulators. In this paper, we revisit the perspective and methodology on constrained optimization paradigms for kinematic control of redundant manipulators.
Published: 18 January 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-10; https://doi.org/10.4108/airo.v1i.6
This paper presents a model-free real-time kinematic tracking controller for a redundant manipulator. Redundant manipulators are common in industrial applications because of the flexibility and dexterity they get from redundant joints. However, at the same time, the modeling of these systems becomes quite challenging, even for simple tasks like trajectory tracking. Some classical approaches are being used to tackle the issue, including a numerical approximation of the Jacobian and pseudo-inverse of the Jacobian matrix. These approaches have their limitations as they require exact parameters for the modeling of the manipulator; they are not immune to position error accumulation with time and put the manipulator way off the target position. Swarm-based meta-heuristic algorithms have given a new direction to the solution of the redundancy resolution problem. However, they are computationally intensive, formulated in discrete-time, and better suited for offline computation rather than real-time. We proposed a novel continuous-time Zeroing Neural Network with Beetle Antennae Search (ZNNBAS). The ZNNBAS algorithm can solve the quadratic optimization problem for redundancy resolution in real-time. To test its performance, we applied it on 7-DOF redundant manipulator with two trajectories to follow: character ``M" and hypotrochoid. The manipulator was able to trace the reference trajectories with minimal tracking errors.
Published: 20 February 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-13; https://doi.org/10.4108/airo.v1i.19
Deep learning is a new area of machine learning research. Deep learning technology applies the nonlinear and advanced transformation of model abstraction into a large database. The latest development shows that deep learning in various fields and greatly contributed to artificial intelligence so far. This article reviews the contributions and new applications of deep learning. The main target of this review is to give the summarize points for scholars to have the analysis about applications and algorithms. Then review tries to investigate the main applications and uses algorithms. In addition, the advantages of using the method of deep learning and its hierarchical and nonlinear functioning are introduced and compared to traditional algorithms in common applications. The following three criteria should be taken into consideration when choosing the area of application. (1) expertise or knowledge of the author; (2) the successful application of deep learning technology has changed the field of application, such as voice recognition, chat robots, search technology and vision; and (3) deep learning can have a significant impact on the application domain and benefit from recent research with natural language and text processing, information recovery and multimodal information processing resulting from multitasking deep learning. This review provides a general overview of a new concept and the growing benefits and popularity of deep learning, which can help researchers and students interested in deep learning methods.
Published: 7 January 2022
Eai Endorsed Transactions on AI and Robotics, Volume 1, pp 1-11; https://doi.org/10.4108/airo.v1i.7
Service robots, e.g. massage robots, have attracted more and more attention in recent years and the most popular study within this field is trajectory tracking. Due to the actual demand for service robots, the solution of trajectory tracking requires fast convergence and high accuracy. In order to solve the above issues, this paper proposed an enhanced Gated recurrent unit (GRU) to deal with trajectory tracking tasks of robot manipulators. The main feature of enhanced GRU is utilizing cell states as well as various gate units to build a novel neural cell. Besides, the presented enhanced GRU resolves the problem of the general neural network model and large memory occupancy. Then the derivations about the computational process of cell state and mixed hidden state of the proposed model have been illustrated. Finally, three trajectory tracking applications, comparison, and visual simulation have verified feasibility as well as the superiority of the enhanced GRU model.