A swarm of robots is the coordination of multiple robots that can perform a collective task and solve a problem more efficiently than a single robot. Over the last decade, this area of research has received significant interest from scientists due to its large field of applications in military or civil, including area exploration, target search and rescue, security and surveillance, agriculture, air defense, area coverage and real-time monitoring, providing wireless services, and delivery of goods. This research domain of collective behaviour draws inspiration from self-organizing systems in nature, such as honey bees, fish schools, social insects, bird flocks, and other social animals. By replicating the same set of interaction rules observed in these natural swarm systems, robot swarms can be created. The deployment of robot swarm or group of intelligent robots in a real-world scenario that can collectively perform a task or solve a problem is still a substantial research challenge. Swarm robots are differentiated from multi-agent robots by specific qualifying criteria, including the presence of at least three agents and the sharing of relative information such as altitude, position, and velocity among all agents. Each agent should be intelligent and follow the same set of interaction rules over the whole network. Also, the system’s stability should not be affected by leaving or disconnecting an agent from a swarm. This survey illustrates swarm systems’ basics and draws some projections from its history to its future. It discusses the important features of swarm robots, simulators, real-world applications, and future ideas.

Deep learning (DL) has been shown to be highly effective in solving various problems across numerous applications and domains, such as autonomous driving and image recognition. Due to the advent of DL, plenty of research works have explored the applicability of DL, more specifically deep neural networks (DNNs), to solve pattern recognition and computer vision challenges. More recently, researchers have focused on the topic of automated generation and exploration of DNN architectures, which tend to mostly focus on image recognition or visual data sets, primarily, due to the computer vision-related DL advancements. In this work, we propose the BioNetExplorer framework to systematically generate and explore multiple DNN architectures for biosignal processing in wearable devices. Our framework varies key neural architecture parameters to search for an embedded DNN architecture with a low hardware overhead, which can be deployed in wearable edge devices to analyze the biosignal data and to extract the relevant information, such as arrhythmia and seizure. Furthermore, BioNetExplorer reduces the exploration time by deploying genetic algorithms, such as NSGA-II, SPEA-2, etc. Our framework also enables the hardware-aware DNN architecture search by imposing user requirements and hardware constraints (storage, FLOPs, etc.) during the exploration stage, thereby limiting the number of networks explored. Moreover, BioNetExplorer can also be used to search for DNNs based on the user-required output classes; for instance, a user might require a specific output class, attributed toward ventricular fibrillation, due to genetic predisposition or a preexisting heart condition. The use of genetic algorithms reduces the exploration time, on average, by $9\times $ , compared to exhaustive exploration. We are successful in identifying Pareto-optimal designs, which can reduce the storage overhead of DNN by $\sim 30$ MB for a quality loss of less than 0.5%. To enable low-cost embedded DNNs, BioNetExplorer also employs different model compression techniques to further reduce the storage overhead of the network by up to $53\times $ for a quality loss of $ < 0.2\%$ .