In this brief, we presented a framework for the trajectory optimization of 5-link Biped Robot using Beetle Antennae Search (BAS) algorithm. The biped robot is highly non-linear and has complex dynamical modeling. It challenging to obtain closed-form solution Robot. modeling two stages, i.e., optimal generation robust control robot. conventional methodologies treat both problems separately are computationally expensive. This brief an problem that combines We employed problem’s...

This paper proposes a model-free control framework for the path planning of rigid and soft robotic manipulator using an intelligent algorithm called Weighted Jacobian Rapidly-exploring Random Tree (WJRRT). The optimization approach is used to model problem, which independent model, then WJRRT solve it. not only explores cartesian space end-effector randomly but also directs it towards goal-position when required. It robust enough tackle uncertainties in make computation more efficient....

Abstract This article proposes a control algorithm for obstacle avoidance and trajectory tracking redundant‐manipulator in smart‐homes. The redundancy provides dexterity flexibility the applications like picking, dropping, transporting objects, predefined paths while avoiding obstacles. is one of critical problems that need to be addressed. Our proposed algorithm, zeroing neural network with beetle antennae search (ZNNBAS), unifies these two into single constrained optimization problem,...

Engineering design optimization problems are difficult to solve because the objective function is often complex, with a mix of continuous and discrete variables various constraints. Our research presents novel hybrid algorithm that integrates benefits sine cosine (SCA) artificial bee colony (ABC) address engineering problems. The SCA recently developed metaheuristic many advantages, such as good search ability reasonable execution time, but it may suffer from premature convergence. enhanced...

The recently emerging multi-portfolio selection problem lacks a proper framework to ensure that client privacy and database secrecy remain intact. Since is of major concern these days, in this paper, we propose variant Beetle Antennae Search (BAS) known as Distributed (DBAS) optimize problems without violating the individual portfolios. DBAS swarm-based optimization algorithm solely shares gradients portfolios among swarm sharing private data or portfolio stock information. hybrid framework,...

In this paper, we address the question of achieving high accuracy in deep learning models for agricultural applications through edge computing devices while considering associated resource constraints. Traditional and state-of-the-art have demonstrated good accuracy, but their practicality as end-user available solutions remains uncertain due to current limitations. One application is detection classification plant diseases image-based crop monitoring. We used publicly PlantVillage dataset...

Accurate leaf segmentation and counting are critical for advancing crop phenotyping improving breeding programs in agriculture. This study evaluates YOLOv11-based models automated detection across spring barley, wheat, winter rye, triticale. The key focus is assessing whether a unified model trained on combined multi-crop dataset can outperform crop-specific models. Results show that the achieves superior performance bounding box tasks, with mAP@50 exceeding 0.85 crops 0.7 crops....

In response to the growing global population and consequent need for sustainable food security, effective pest management is critical enhancing agricultural productivity. This research presents YOLOv8, a state-of-the-art deep learning model optimized detection in environments, contributing modern security efforts. Evaluated using complex IP102 dataset, YOLOv8 demonstrated notable improvements accuracy, achieving scores of 66.9 mAP@0.5 42.1 mAP@[0.5:0.95]. These results underscore YOLOv8’s...

The issue of inventory balance in supply chain management represents a classic problem within the realms and logistics. It can be modeled using mixture equality inequality constraints, encompassing specific considerations such as production, transportation, limitations. A Zeroing Neural Network (ZNN) model for time-varying linear equations systems is presented this manuscript. In order to convert these into mixed nonlinear framework, method entails adding non-negative slack variable. ZNN...

This paper presents a model-free real-time kinematic tracking controller for redundant manipulator. Redundant manipulators are common in industrial applications because of the flexibility and dexterity they get from joints. However, at same time, modeling these systems becomes quite challenging, even simple tasks like trajectory tracking. Some classical approaches being used to tackle issue, including numerical approximation Jacobian pseudo-inverse matrix. These have their limitations as...

In this paper, we presented an autonomous control framework for the wall following robot using optimally configured Gated Recurrent Unit (GRU) model with hyperband algorithm. GRU is popularly known time-series or sequence data, and it overcomes vanishing gradient problem of RNN. also consumes less memory computationally more efficient than LSTMs. The selection hyper-parameters a complex optimization local minima. Usually, are selected through hit trial, which does not guarantee optimal...

<title>Abstract</title> Detecting and quantifying the diseased regions in a leaf is an important task plant breeding order to select plants based on disease resistance. Ratings done by eye hand are not accurate, can be highly subjective take lot of work hours. Using machine learning, it possible generate faster more accurate data. By combining modern anomaly detection algorithms with masking robust method capable estimating infected area was developed, resulting novel superior results....