Inverse kinematics involves calculating joint configurations for a robot manipulator to perform tasks. To solve this problem, various approaches have been proposed, including meta-heuristics methods, which are not specific to a manipulator class and do not require matrix inversion. Particle Swarm Optimization (PSO) is a commonly used meta-heuristic due to its simplicity and effectiveness, but it is prone to local minima. In response to this challenge, several PSO variants have been introduced, incorporating strategies such as multiple populations, adaptive parameters, and particle resampling. This paper focuses on the variant Fully Resample PSO (FRPSO), which performs full particle resampling in each iteration. The study investigates how the number of joints and FRPSO parameters impact its performance. Additionally, the paper proposes a novel extension to FRPSO called Modified-FRPSO. The novel approach involves one less parameter compared to the original method, simplifying the process of fine-tuning the performance of FRPSO. Furthermore, the experiments conducted in environments with obstacles indicate that the new approach outperforms the original FRPSO and PSO in terms of convergence rate and average number of iterations.

This paper explores a data augmentation approach for images of rigid bodies, particularly focusing on electrical equipment and analogous industrial objects. By leveraging manufacturer-provided datasheets containing precise equipment dimensions, we employed straightforward algorithms to generate synthetic images, permitting the expansion of the training dataset from a potentially unlimited viewpoint. In scenarios lacking genuine target images, we conducted a case study using two well-known detectors, representing two machine-learning paradigms: the Viola–Jones (VJ) and You Only Look Once (YOLO) detectors, trained exclusively on datasets featuring synthetic images as the positive examples of the target equipment, namely lightning rods and potential transformers. Performances of both detectors were assessed using real images in both visible and infrared spectra. YOLO consistently demonstrates $F_1$ scores below 26% in both spectra, while VJ’s scores lie in the interval from 38% to 61%. This performance discrepancy is discussed in view of paradigms’ strengths and weaknesses, whereas the relatively high scores of at least one detector are taken as empirical evidence in favor of the proposed data augmentation approach.

Particle Swarm Optimization (PSO) is a widely used solver for the Inverse Kinematics (IK) problem, offering competitive convergence rates compared to methods like Damped Least Squares and Forward and Backward Reaching Inverse Kinematics. However, PSO’s high computational cost per iteration remains a challenge. While many PSO variants focus on reducing the number of iterations, this work introduces a novel approach to decrease the computational cost per iteration by minimizing direct kinematics calculations. Leveraging quantum computing algorithms, specifically Quantum Amplitude Amplification (QAA), we developed Quantum-Grover PSO (QG-PSO), which identifies a suboptimal particle to guide the swarm. This approach capitalizes on QAA’s quadratic speedup for unstructured search problems. Simulated experiments demonstrate that QG-PSO significantly reduces computational costs compared to classical PSO when solving the IK problem for robotic manipulators with 7 and 15 joints, highlighting the potential of quantum computing in robotics. In addition to position correction, we considered orientation correction while respecting joint limits and avoiding obstacles.

The changes are described below