Abstract
Nowadays, there has been a growing interest in the use of mobile robots for various applications, where the analysis of the operational environment is a crucial component to conduct our special tasks or missions. The main aim of this work was to implement artificial intelligence (AI) for object detection and distance estimation navigating the developed unmanned platform in unknown environments. Conventional approaches are based on vision systems analysis using neural networks for object detection, classification, and distance estimation. Unfortunately, in the case of precise operation, the used algorithms do not provide accurate data required by platforms operators as well as autonomy subsystems. To overcome this limitation, the authors propose a novel approach using the spatial data from laser scanners supplementing the acquisition of precise information about the detected object distance in the operational environment. In this article, we introduced the application of pretrained neural network models, typically used for vision systems, in analysing flat distributions of LiDAR point cloud surfaces. To achieve our goal, we have developed software that fuses detection algorithm (based on YOLO network) to detect objects and estimate their distances using the MiDaS depth model. Initially, the accuracy of distance estimation was evaluated through video stream testing in various scenarios. Furthermore, we have incorporated data from a laser scanner into the software, enabling precise distance measurements of the detected objects. The paper provides discussion on conducted experiments, obtained results, and implementation to improve performance of the described modular mobile platform.

Abstract
Forecasting methods are affected by data quality issues in two ways: 1. they are hard to predict, and 2. they may affect the model negatively when it is updated with new data. The latter issue is usually addressed by pre-processing the data to remove those issues. An alternative approach has recently been proposed, Corrector LSTM (cLSTM), which is a Read & Write Machine Learning (RW-ML) algorithm that changes the data while learning to improve its predictions. Despite promising results being reported, cLSTM is computationally expensive, as it uses a meta-learner to monitor the hidden states of the LSTM. We propose a new RW-ML algorithm, Kernel Corrector LSTM (KcLSTM), that replaces the meta-learner of cLSTM with a simpler method: Kernel Smoothing. We empirically evaluate the forecasting accuracy and the training time of the new algorithm and compare it with cLSTM and LSTM. Results indicate that it is able to decrease the training time while maintaining a competitive forecasting accuracy.

Abstract

Abstract
Aquaculture presents itself as one of the most rapidly developing means of sustainable production of animal protein to feed ever-growing populations. Recirculating aquaculture systems offer higher control and fewer inconveniences than traditional systems, making them an attractive option for fish production. Although the sector's digitalization is in its early stages, its application should increase its rentability while conserving the environment. This paper aims to promote the sector's evolution by assessing parameter importance in mortality with tree-based machine learning models, verifying the method's natural robustness and how it compares to a specially devised one, and at the same time evaluating the concept's relevance in predicting categorical mortality values. In particular, to better understand the aquaculture production process through a systematic data evaluation, an exploration based on real-time data acquisition is fully needed. Moreover, algorithm robustness is a key ingredient in this application since measurements are greatly affected by errors. This invalidates the application of traditional machine learning methods, where models are sensitive to production data variations and sensor noise. The study found the parameters that play relevant roles in the production phases, such as pH and nitrate concentration. While the obtained predictive metrics are still sub-optimal, further enhancements could be achieved through rigorous analysis of feature engineering, fine-tuning model hyperparameters, and exploring more advanced algorithms. Additionally, incorporating larger and more diverse datasets, refining data pre-processing techniques, and iteratively optimizing the model architecture may contribute to significant improvements in predictive performance. Despite that, the impact costs of using adjusted machine learning metrics are clear, as are the importance of data rounding in pre-processing and directions for improvement regarding data acquisition and transformation.

Abstract
The increasing development of Unmanned Surface Vehicles (USVs) for various applications in open and shallow waters has increased demand for more advanced USVs with improved safety and navigation systems. This article introduces a collision avoidance system for USVs that complies with the International Regulations for Preventing Collisions at Sea (COLREG) rules, particularly rules 13 to 18 from Part B - Steering and Sailing. The system utilizes a three-block architecture for risk assessment, situation identification, and path replanning. Practical testing and validation were conducted using the Stonefish simulator, demonstrating the system's effectiveness in ensuring compliance with COLREG rules and facilitating safe navigation of USVs.

Abstract