The underwater industry and scientific community are actively researching the development of vehicles that combine the functionalities of autonomous underwater vehicles and remotely operated vehicles. An innovative approach to address the challenges posed by underwater exploration is the development of autonomous underwater reconfigurable vehicles (AURVs). These vehicles are designed to adapt their configuration to suit the requirements of the task at hand. The flexibility of AURVs enables them to undertake a variety of underwater missions, ranging from scientific research to deep-sea exploration. The Department of Industrial Engineering at the University of Florence, Italy, has developed and patented an innovative AURV that is able to quickly change its shape to suit different tasks. The reconfigurable underwater vehicle for inspection, free-floating intervention and survey tasks (RUVIFIST) have been equipped with two extreme configurations. The first configuration is a slender one meant for long navigation tasks, while the second configuration is a stocky one designed for tackling complex objectives such as inspection or intervention operations. With the ability to adapt its form to suit the task at hand, the RUVIFIST vehicle represents a significant advancement in underwater vehicle technology. This work provides an overview of the challenges faced and the solutions adopted during the development of this new vehicle. This article presents the results of experimental campaigns to test the reconfigurable system of the vehicle and the strategies developed for the guidance, navigation, and control system of AURVs. Finally, preliminary tests were conducted to explore the integration of machine learning and deep learning algorithms that are compatible with the purpose of automatic target recognition.

In recent years, the number of people with disabilities has increased hugely, especially in low- and middle-income countries. At the same time, robotics has made significant advances in the medical field, and many research groups have begun to develop low-cost wearable solutions. The Mechatronics and Dynamic Modelling Lab of the Department of Industrial Engineering at the University of Florence has recently developed a new version of a wearable hand exoskeleton for assistive purposes. In this paper, we will present a new regression method to predict the finger angle position of the first joint from the value of the sEMG of the forearm and the previous position of the finger itself. To acquire the dataset necessary to train the regressor a specific graphical user interface was developed which was able to acquire sEMG data from a Myo armband and the finger position from a Leap Motion Controller. Two long short-term memory (LSTM) models were compared, one in its standard configuration and the other with a convolutional layer, yielding significantly better performance for the second one, with an increase in $R^2$ coefficient from an average value of $0.746$ to $0.825$, leading to the conclusion that a convolutional layer could increase performance when few sensors are available.