Tracking people has many applications, such as security or safe use of robots. Many onboard systems are based on Laser Imaging Detection and Ranging (LIDAR) sensors. Tracking peoples' legs using only information from a 2D LIDAR scanner in a mobile robot is a challenging problem because many legs can be present in an indoor environment, there are frequent occlusions and self-occlusions, many items in the environment such as table legs or columns could resemble legs as a result of the limited information provided by two-dimensional LIDAR usually mounted at knee height in mobile robots, etc. On the other hand, LIDAR sensors are affordable in terms of the acquisition price and processing requirements. In this article, we describe a tool named PeTra based on an off-line trained full Convolutional Neural Network capable of tracking pairs of legs in a cluttered environment. We describe the characteristics of the system proposed and evaluate its accuracy using a dataset from a public repository. Results show that PeTra provides better accuracy than Leg Detector (LD), the standard solution for Robot Operating System (ROS)-based robots.

Robots and autonomous systems in general are set to suffer similar cybersecurity problems that computers have been facing for decades. This is not only worrying for critical tasks such as those performed by surgical, or military robots but also for household robots such as vacuum cleaners or for teleconference robots compromise privacy and safety of their owners. What will happen if these robots are hacked? This study presents a survey on the cybersecurity attacks associated with service robots, and as a result, a taxonomy that classifies the risks faced by users when using service robots, distinguishing between security and safety threads, is presented. We also present the robot software development phase as one the most relevant ones for the security of robots.

Version control systems’ usage is a highly demanded skill in information and communication technology professionals. Thus, their usage should be encouraged by educational institutions. This work demonstrates that it is possible to assess if a student can pass a computer science-related subject by monitoring its interaction with a version control system. This paper proposes a methodology that compares the performance of several machine learning models so as to select the appropriate predicting model for the assessment of the students’ achievements. To fit predicting models, three subjects of the Degree in Computer Science at the University of León are considered to obtain the dataset: computer organization, computer programming, and operating systems extension. The common aspect of these subjects is their assignments, which are based on developing one or several programs with programming languages such as C or Java. To monitor the practical assignments and individual performance, a Git repository is employed allowing students to store source code, documentation, and supporting control versions. According to the presented experience, there is a huge correlation between the level of interaction for each student and the achieved grades.