Autonomous underwater vehicles (AUVs) have become attractive and essential for underwater search and exploration because of the advantages they offer over manned underwater vehicles. Hence the need to improve AUV technologies. One crucial area of AUV technology involves efficiently solving the path planning problem. Several approaches have been identified from the literature for AUV global and local path planning. The use of machine learning (ML) techniques in overcoming some of the challenges associated with AUV path planning problems such as safety and obstacle avoidance, energy consumption, and optimal time and distance travelled remains an active research area. While there is literature on global and local path planning that explores different techniques, there is still a lack of paper that provides an overview of the application of ML for local path planning. Hence the main objective of this paper is to present an overview of the state-of-the-art application of ML techniques on local path planning for AUVs. The ML algorithms are discussed under supervised, unsupervised, and reinforcement learning. The challenges faced in real-life deployment, simulated scenarios, computational issues, and application of ML algorithms are discussed, with future research directions presented.

Due to the significant advances of wireless sensor networks (WSNs), researchers are eager to use this technology in the subsea applications. Because of rapid absorption of high radio frequency in the water, acoustic waves are used as communication medium, which pose new challenges, including high propagation delay, high path loss, low bandwidth, and high-energy consumption. Because of these challenges and high movement of nodes by water flow, end-to-end routing methods used in most of existing routing protocols in WSNs are not applicable to underwater environments. Therefore, new routing protocols have been developed for underwater acoustic sensor networks (UWASNs) in which most of the routing protocols take advantage of greedy routing. Due to inapplicability of global positioning system (GPS) in underwater environments, finding location information of nodes is too costly. Therefore, based on a need for location information, we divided the existing greedy routing protocols into two distinctive categories, namely, location-based and location-free protocols. In addition, location-free category is divided into two subcategories based on method of collecting essential information for greedy routing, including beacon-based and pressure-based protocols. Furthermore, a number of famous routing protocols belonging to each category are reviewed, and their advantages and disadvantages are discussed. Finally, these protocols are compared with each other based on their features.

Mobile Ad Hoc Networks (MANETs) are dynamic networks without fixed infrastructure, making them particularly vulnerable to security threats such as black and gray hole attacks. As these attacks grow more sophisticated, advancing detection methods become critical. This survey critically evaluates existing detection techniques and identifies major gaps in current research. It focuses on a comprehensive classification and in-depth analysis of attack detection methods, particularly those employing advanced Machine Learning techniques. We adopt a structured approach, analyzing MANET characteristics and detailing black and gray hole attacks. The evaluation covers various detection and mitigation techniques, with a strong emphasis on the innovative use of ML and optimization methods like Federated Learning (FL), reinforcement learning, and metaheuristic algorithms. Our findings indicate that advanced ML techniques, especially Long Short-Term Memory (LSTM) networks and FL, significantly enhance detection accuracy and robustness against these attacks. We also discussed the potential of game theory and reinforcement learning for optimizing routing protocols and improving network resilience. The survey underscores the necessity for ongoing research into more sophisticated and adaptable detection mechanisms, urging both academic and practical communities to explore novel approaches for developing more secure, efficient MANET systems.

Mobile ad hoc networks (MANETs) face significant challenges in maintaining secure and efficient communication owing to their dynamic nature and vulnerability to security threats. Traditional routing protocols often struggle to adapt to rapidly changing topologies and potential malicious nodes, compromising network performance and security. This study addresses these challenges by proposing FLSTMT-LAR (Federated Learning Long Short-Term Memory Trust-aware Location-aided Routing), a novel framework that integrates multiobjective optimization with LSTM-based trust prediction for robust routing decisions, implements a decentralized federated learning mechanism for collaborative trust model updates while preserving node privacy, incorporates dynamic trust assessment using LSTM networks for accurate temporal behavior pattern analysis, and provides an adaptive routing decision mechanism that effectively balances multiple performance objectives including trustworthiness, energy efficiency, and network latency. We evaluate this framework against existing protocols across various scenarios, including different network densities, mobility patterns, and malicious node percentages. Results demonstrate FLSTMT-LAR’s superior performance in high-threat environments, achieving up to 80% packet delivery ratio compared with 45% for traditional approaches. In mobile scenarios, it shows improved adaptability, maintaining consistent performance as network density increases. MOO, particularly nondominated sorting genetic algorithm III, effectively balances conflicting network objectives, offering a 15% improvement in overall network performance compared with single-objective approaches. These findings highlight the potential of integrating advanced machine learning and optimization techniques in MANET routing protocols, paving the way for secure, efficient, and adaptive network communications in challenging environments.

A new approach to robot motion planning is proposed by applying ant colony optimization with the probabilistic roadmap planner (PRM). The PRM is a path planning method that consists of capturing the connectivity of the robot's free space in a network called the roadmap. An ant colony robot motion planning (ACRMP) method is proposed that takes the benefit of collective behaviour of ants foraging from a nest to a food source. Two groups of ants are placed at both the nest and food source respectively. A number of ants (agents) are released from the nest (start configuration) and begin to forage (search) towards the food (goal configuration). Each ant has a certain quantity of pheromone to be dropped along the path. The ants track down the pheromone trails previously dropped by the nest's ants to accomplish the path between the two points of nest and food respectively. Results from preliminary tests show that the ACRMP is capable of reducing the intermediate configuration between the initial and goal configuration in an acceptable running time