In this paper, the mobile robot motion planning approach (PSO-CF2-mt: PSO-CF2 for moving targets) is improved to track a moving target following a smooth path. PSO-CF2-mt was tested previously in various static and dynamic environments and proved its capacity to track moving targets whatever the form of the target’s trajectory. The problem with PSO-CF2-mt that we want to face in this paper is the stability of the robot’s motion. It is tackled by limiting the angular velocity of the robot. The angular speed limit is PSO selected and the variance of the angular speed is added to the fitness function as a third objective. This new version of PSO-CF2-mt is called OS-CF2-mt (Optimized Stable CF2-mt). Simulation results prove the capacity of OS-CF2-mt in ensuring stable travels for mobile robots following short, secure and smooth paths when tracking moving targets whether the environment is static or dynamic and whatever the form of the target’s trajectory.

This study underscores the critical role of accurately classifying brain tumors in Magnetic Resonance Imaging (MRI) for clinical diagnosis and treatment planning. Deep Transfer Learning (DTL) emerges as a potent method for image classification, utilizing pre-trained models to extract meaningful features. However, the inherent complexity of DTL models often obscures their decision-making processes, thereby challenging the trust of medical experts. To address this, the study introduces an Explainable Artificial Intelligence (XAI)-enhanced framework that improves the interpretability of DTL models. This framework employs advanced XAI techniques to visually elucidate their decision logic, utilizing well-established pre-trained models such as Xception, ResNet50-V2, DenseNet121, and EfficientNetB0-V2 from the ImageNet dataset. Using a dataset comprising three primary categories of brain tumors (gliomas, meningiomas, and pituitary tumors) alongside a category indicating the absence of tumors. The accuracy findings of the proposed framework are encouraging. which increases the interest of radiologists in its utility as a second opinion in clinical diagnosis. Finally, eXplainable Artificial Intelligence (XAI) approaches, including LIME, SHAP, and GRADCAM, are employed to elucidate the results, with GRAD-CAM identified as the most effective explanatory method among them.

Autonomous$\text{PSO-DVSF}^{\text{2}}$is a PSO optimized$\mathrm{F}^{2}$based mobile robot motion planning approach that we have previously proposed to guide a robot in unknown environments. We proved the efficiency of this approach by means of simulation tests [15]. Hence in this paper, we are dealing with the experimental setup of our proposed simulations. As a first step, the approach is tested in the MobileSim virtual environment then in a second step the tests are done in experimental environments using the differential Pioneer P3-DX wheeled robot. The results of these tests proved the real efficiency of autonomous PSO-$\text{DVSF}^{\text{2}}$algorithm to reach its destination selecting the shortest and the securest trajectory whatever the environment is static or dynamic and whatever its complexity.

The results of the Deep Learning (DL) are indisputable in different fields and in particular that of the medical diagnosis. The black box nature of this tool has left the doctors very cautious with regard to its estimates. The eXplainable Artificial Intelligence (XAI) recently seemed to lift this challenge by providing explanations to the DL estimates. Several works are published in the literature offering explanatory methods. We are interested in this survey to present an overview on the application of XAI in Deep Learning-based Magnetic Resonance Imaging (MRI) image analysis for Brain Tumor (BT) diagnosis. In this survey, we divide these XAI methods into four groups, the group of the intrinsic methods and three groups of post-hoc methods which are the activation based, the gradientr based and the perturbation based XAI methods. These XAI tools improved the confidence on the DL based brain tumor diagnosis.

This paper presents the PSO-CF2-mt motion planning approach that we propose for two wheeled mobile robots for tracking moving targets in known dynamic environments. The Particle Swarm Optimization Canonical Force Field (PSO-CF2) is a mobile robot motion planning approach that we have previously proposed for static and dynamic environments[9]. PSO-CF2-mt is an improved version of PSO-CF2to become capable of tracking a moving target. The basic concept of PSO-CF2-mt is to generate a continually changing parameterized Force Field for the robot based on the characteristics of all objects presents in the environment. The simulation results prove clearly the ability of PSO-CF2-mt to follow and reach a moving target by choosing the shortest and the most secure paths whatever the complexity of the environment and the form of the target's trajectory. A comparative study with APF (Artificial Potential Field) proves the quality of our proposed approach.

The important role that mammography is playing in breast cancer detection can be attributed largely to the technical improvements and dedication of radiologists to breast imaging. A lot of work is being done to ensure that these diagnosing steps are becoming smoother, faster and more accurate in classifying whether the abnormalities seen in mammogram images are benign or malignant. In this paper, an evolutionary approach for design of TSK-type fuzzy model (TFM) is proposed to solve the breast cancer diagnosis problem. In the proposed method, both the number of fuzzy rules and adjustable parameters in the TFM are designed concurrently combining the compact genetic algorithm (CGA) and the steady-state genetic algorithm (SSGA). The computational experiments show that the presented approach can obtain better generalization than some existing methods reported recently in the literature using the widely accepted Wisconsin breast cancer diagnosis (WBCD) database.