Multi-robot systems (MRS) are a group of robots that are designed aiming to perform some collective behavior. By this collective behavior, some goals that are impossible for a single robot to achieve become feasible and attainable. There are several foreseen benefits of MRS compared to single robot systems such as the increased ability to resolve task complexity, increasing performance, reliability and simplicity in design. These benefits have attracted many researchers from academia and industry to investigate how to design and develop robust versatile MRS by solving a number of challenging problems such as complex task allocation, group formation, cooperative object detection and tracking, communication relaying and self-organization to name just a few. One of the most challenging problems of MRS is how to optimally assign a set of robots to a set of tasks in such a way that optimizes the overall system performance subject to a set of constraints. This problem is known as Multi-robot Task Allocation (MRTA) problem. MRTA is a complex problem especially when it comes to heterogeneous unreliable robots equipped with different capabilities that are required to perform various tasks with different requirements and constraints in an optimal way. This chapter provides a comprehensive review on challenging aspects of MRTA problem, recent approaches to tackle this problem and the future directions.

Imitation learning techniques aim to mimic human behavior in a given task. An agent (a learning machine) is trained to perform a task from demonstrations by learning a mapping between observations and actions. The idea of teaching by imitation has been around for many years; however, the field is gaining attention recently due to advances in computing and sensing as well as rising demand for intelligent applications. The paradigm of learning by imitation is gaining popularity because it facilitates teaching complex tasks with minimal expert knowledge of the tasks. Generic imitation learning methods could potentially reduce the problem of teaching a task to that of providing demonstrations, without the need for explicit programming or designing reward functions specific to the task. Modern sensors are able to collect and transmit high volumes of data rapidly, and processors with high computational power allow fast processing that maps the sensory data to actions in a timely manner. This opens the door for many potential AI applications that require real-time perception and reaction such as humanoid robots, self-driving vehicles, human computer interaction, and computer games, to name a few. However, specialized algorithms are needed to effectively and robustly learn models as learning by imitation poses its own set of challenges. In this article, we survey imitation learning methods and present design options in different steps of the learning process. We introduce a background and motivation for the field as well as highlight challenges specific to the imitation problem. Methods for designing and evaluating imitation learning tasks are categorized and reviewed. Special attention is given to learning methods in robotics and games as these domains are the most popular in the literature and provide a wide array of problems and methodologies. We extensively discuss combining imitation learning approaches using different sources and methods, as well as incorporating other motion learning methods to enhance imitation. We also discuss the potential impact on industry, present major applications, and highlight current and future research directions.

There is a pressing need for novel and innovative therapeutic strategies to address infections caused by intracellular pathogens. Peptide nucleic acids (PNAs) present a novel method to target intracellular pathogens due to their unique mechanism of action and their ability to be conjugated to cell penetrating peptides (CPP) to overcome challenging delivery barriers. In this study, we targeted the RNA polymerase α subunit (rpoA) using a PNA that was covalently conjugated to five different CPPs. Changing the conjugated CPP resulted in a pronounced improvement in the antibacterial activity observed against Listeria monocytogenes in vitro, in cell culture, and in a Caenorhabditis elegans (C. elegans) infection model. Additionally, a time-kill assay revealed three conjugated CPPs rapidly kill Listeria within 20 minutes without disrupting the bacterial cell membrane. Moreover, rpoA gene silencing resulted in suppression of its message as well as reduced expression of other critical virulence genes (Listeriolysin O, and two phospholipases plcA and plcB) in a concentration-dependent manner. Furthermore, PNA-inhibition of bacterial protein synthesis was selective and did not adversely affect mitochondrial protein synthesis. This study provides a foundation for improving and developing PNAs conjugated to CPPs to better target intracellular pathogens.

The aim of this work is to integrate and analyze the performance of a path planning method based on Time Elastic Bands (TEB) in real research platform based on Ackermann model. Moreover, it will be proved that all modules related to the navigation can coexist and work together to achieve the goal point without any collision. The study is done by analyzing the trajectory generated from global and local planners. The software prototyping tool is Robot Operating System (ROS) from Open Source Robotics Foundation and the research platform is the iCab (Intelligent Campus Automobile) from University Carlos III. This work has been validated from a test inside the campus where the iCab has performed the navigation between the starting point and the goal point without any collision. During the experiment, we proved the low sensitivity of the TEB method to variations of the vehicle model configuration and constraints.

No abstract