A Comprehensive Review of Bio-Inspired Approaches to Coordination, Communication, and System Architecture in Underwater Swarm Robotics

AI summary

Overview

Research area: Underwater swarm robotics — specifically bio-inspired coordination and optimisation algorithms, underwater communication modalities, and system/hardware architecture for multi-vehicle marine operations.

Technical level: Intermediate. The paper is a review rather than a new algorithmic result; it assumes familiarity with swarm robotics concepts but defines them, and it tabulates concrete ranges, data rates, and algorithm families that a newcomer can follow.

Scope (one sentence): The review synthesises 446 peer-reviewed articles (2001–2025) covering marine bio-inspired algorithms, underwater communication constraints, and system design, and organises them under a four-dimensional classification framework spanning communication dependency, environmental adaptability, energy efficiency, and swarm scalability.

What This Paper Is About

Underwater robotics research is split across three communities that rarely talk to each other: people designing bio-inspired coordination algorithms, people working on underwater communication links, and people building the hardware. The result is that algorithms are proposed without regard for the bandwidth, latency, and energy limits that decide whether they can actually run on a real swarm. This review pulls those three strands into a single comparative framework, using four marine-specific algorithms — the Artificial Fish Swarm Algorithm (AFSA), Whale Optimisation Algorithm (WOA), Coral Reef Optimisation (CRO), and Marine Predators Algorithm (MPA) — as the running exemplars of how biological behaviour is translated into coordination logic.

Key Contributions

1. Three-layer unification. The review explicitly couples bio-inspired coordination and optimisation, underwater communication constraints, and system architecture, so that algorithms are discussed alongside the physical and networking limits that shape real deployments.
2. Marine-grounded algorithm survey. It focuses on bio-inspired methods derived specifically from marine organisms and their underwater adaptations, using AFSA, WOA, CRO, and MPA as representative exemplars to illustrate common design patterns and evaluation considerations.
3. Four-dimensional classification framework. It introduces a framework based on communication dependency, environmental adaptability, energy efficiency, and swarm scalability to position and compare approaches consistently across the corpus.
4. Evaluation-gap synthesis. It consolidates recurring evaluation gaps and open challenges, with the aim of supporting more reproducible benchmarking and more informed algorithm selection.

The authors also position the work against existing surveys: unlike surveys that treat bio-inspired algorithms broadly across general robotics, or that concentrate on narrow subsets of underwater coordination techniques, this one is explicitly underwater-specific and integrates algorithms, communications, and architecture into a single comparative framework.

Main Findings

• Field growth: A publication-trend analysis covering 2001 to 2025, based on 446 research articles from major databases, shows exponential growth in underwater swarm robotics research.
• Corpus composition: The final corpus is 446 peer-reviewed research articles. Roughly 65% of papers from the original sources were excluded during filtering and consolidation, removing review papers, conference papers, and duplicate publications.
• GPS denial dominates design: GPS signals do not propagate underwater, forcing AUVs onto Inertial Navigation Systems, dead reckoning, and acoustic positioning arrays (Long Baseline and Ultra-Short Baseline). INS accumulates drift over time, dead reckoning suffers from bias and integration errors, and acoustic positioning needs pre-deployed infrastructure and is constrained by propagation delays and multipath interference. The consequence is that individual agents may hold inconsistent position estimates relative to their neighbours, which complicates formation control.
• Communication trade-offs are stark: Acoustic links give the longest range (roughly 100 m to 20 km) at 10 bps to 100 kbps and about 1500 m/s propagation, but with high sensitivity to temperature and salinity. Optical links reach 1 Mbps to 1 Gbps over 1 to 100 m but are very sensitive to turbidity and scattering; clear ocean water allows 60 to 70 m while turbid water reduces it below 10 m. Electromagnetic/RF links are confined to under 2 m in water. Magnetic Induction operates over short-to-medium ranges with negligible propagation delay and low sensitivity to water properties, and is proposed for local coordination. Hybrid acoustic–optical–MI architectures are proposed to balance these trade-offs via adaptive switching protocols.
• Acoustic remains the bottleneck for large swarms: Acoustic channels are limited to roughly 10 kbaud data rates and suffer high latency and noise, which the review identifies as making coordination challenging in large swarms.
• Energy-efficient schooling has a biological precedent: Fish reduce energy expenditure per tail beat by more than 50% compared with solitary individuals through hydrodynamic interactions, and fish schools maintain cohesion through local alignment and spacing rather than global control. Hydrodynamic coupling between individuals can reduce overall energy expenditure by more than 50% compared with solitary swimming.
• Coordination can be extremely minimal: Binary coordination strategies in which AUVs make left-or-right rotation decisions based solely on neighbour headings achieve coherent collective motion with no explicit synchronisation and no global positioning.
• CRO encodes competition as task allocation: CRO operates on a discretised reef grid of cells storing candidate solutions. Larvae generated from parent solutions (via crossover plus random injection) attempt to settle into a randomly chosen cell; a larva replaces the resident only if its fitness is higher, implementing local competition for space. Periodic depredation removes a fraction of the weakest corals with probability P_d, preventing stagnation.
• MPA encodes role switching: MPA agents act as explorers at high predator–prey velocity ratios and as refiners at low velocity ratios, giving a phase-based transition that the review describes as adaptive task allocation responding to mission progress.
• Validation is almost entirely simulated: Every hybrid algorithm listed in the review's comparison table reports verification through simulation only. These include an IMOPSO–DWA hybrid for 3D cooperative path planning (multiple robots, 2025), a neural leader–follower control scheme for 2D-with-depth formation control (5 robots: 1 leader plus 4 followers, 2025), a hybrid leader–follower formation controller in 3D (7 robots: 1 leader plus 6 followers, 2023), a Fast Graph Pointer Network for task assignment in 2D (multiple robots, 2023), a PSO + ELKAI solver for 3D coverage path planning (multiple robots, 2025), and a hybrid artificial potential field method (HAP–IAPF) for cooperative hunting in 3D (8 robots, 2022).
• Note on completeness: The supplied text ends partway through Section 4.3 on bio-inspired communication protocols. The detailed contents of Section 5 (system design and implementation), Section 6 (the full integrative classification and comparative evaluation), and the paper's consolidated future research directions are referenced in the abstract and in cross-references but are not present in the provided content, so their specific findings are not reported here.

Methodology in Plain English

The authors ran a structured narrative review rather than a straight systematic review, guided by a three-phase design that combines scoping review principles, thematic analysis, and conceptual framework development. They searched seven scholarly databases — MDPI, OpenAlex, Scopus, Springer, ScienceDirect, IEEE, and Google Scholar — using Boolean keyword combinations such as ("underwater" OR "marine") AND ("swarm" OR "multi-robot" OR "multi-AUV") AND ("robotics" OR "coordination" OR "cooperation"), plus specific phrases like "marine swarm robotics", "AUV swarm robotics", and "underwater swarm robotics". Searches were restricted to English-language publications dated 2001 to 2025, and only peer-reviewed research articles were kept; review papers, conference proceedings, theses, and non-peer-reviewed sources were excluded. Duplicates were removed by matching DOIs and comparing titles. Each surviving article was assigned to one or more thematic categories — bio-inspired coordination and control algorithms; underwater communication and networking strategies; system design and implementation (including platforms, sensing, energy management, and validation) — and that coding drives the review's structure. The authors state plainly that they do not claim precise percentage coverage of all relevant publications, because indexing overlaps and varies across venues, and that the dataset may have gaps and the search strategy may miss work using alternative terminology.

Why This Matters

Impact on research. The paper's core diagnosis is that the field is conceptually and methodologically fragmented: substantial work exists on swarm coordination, communication, and bio-inspired design, but no unified framework integrates these dimensions into a coherent body of knowledge. By supplying a four-dimensional comparison axis and by flagging that essentially all the surveyed hybrid coordination methods were verified only in simulation, it gives the community a shared vocabulary for comparison and a concrete argument for hardware validation.

Real-world applications.

• Seabed mapping and habitat and water-quality monitoring, where multiple vehicles improve spatial coverage.
• Offshore infrastructure inspection, including pipeline inspection and relay tasks that the review associates with linear-chain topologies.
• Search-and-recovery operations, where swarm redundancy provides mission resilience.
• Operations in remote environments such as under-ice and deep-sea regions, where communication is sparse, delayed, and intermittent.
• Short-range high-data-rate tasks such as docking or sensor payload transmission, where optical links' lower energy per bit makes them attractive.

Industry relevance. The review frames underwater swarm robotics as a response to the growing complexity of marine operations and the sustainable management of marine resources. The practical payoff it points to is collective autonomy — moving from single-vehicle autonomy to cooperation that enhances spatial coverage, robustness, and mission adaptability. The communication analysis is directly relevant to anyone specifying hardware: acoustic versus optical versus magnetic induction versus hybrid choices set the bandwidth, latency, and range envelope that determines feasible neighbourhood sizes, update rates, and degree of decentralisation.

Future Directions

• Hardware validation over simulation. The review notes that the hybrid coordination methods it tabulates are verified in simulation only and identifies recurring evaluation gaps; closing them requires real deployments under realistic acoustic and optical connectivity.
• Integrating learning with marine bio-inspired algorithms. Most marine-specific bio-inspired algorithms employ fixed behavioural rules. The review points to hybrid approaches that combine reinforcement learning and adaptive parameter tuning as a route to greater responsiveness, citing neural network-based Q-learning for path planning, spinal neural system models for heterogeneous AUV cooperative hunting, and neuromodulation for motor control.
• Cross-layer design. The review argues for tighter coupling between algorithmic, communication, and hardware perspectives so that biologically inspired rules are matched to the neighbourhood sizes, update rates, and switching behaviour that hybrid acoustic–optical–MI networks can actually support.
• More reproducible benchmarking. The stated fourth contribution — synthesising evaluation gaps — implies that the field needs consistent benchmarking practice before algorithm selection can be informed rather than anecdotal.
• Adaptive coordination under degraded links. Adaptive strategies that adjust parameters in response to varying communication quality, and decision propagation through local neighbour interactions rather than global synchronisation, are identified as key directions for future marine robotic swarms.

Target Audience

Researchers and graduate students entering underwater or marine swarm robotics who need a single entry point spanning algorithms, communications, and hardware; robotics and communications engineers specifying AUV, ROV, or USV multi-vehicle systems; and reviewers or programme planners who need a comparative framework rather than a single-technique survey. It is written at an intermediate level: accessible to readers with basic swarm robotics or marine systems background, and useful as a structured reference for experts who want the cross-layer view rather than algorithmic detail alone.

Authors’ abstract

The increasing complexity of marine operations has intensified the need for intelligent robotic systems to support ocean observation, exploration, and resource management. Underwater swarm robotics offers a promising framework that extends the capabilities of individual autonomous platforms through collective coordination. Inspired by natural systems, such as fish schools and insect colonies, bio-inspired swarm approaches enable distributed decision-making, adaptability, and resilience under challenging marine conditions. Yet research in this field remains fragmented, with limited integration across algorithmic, communication, and hardware design perspectives. This review synthesises bio-inspired coordination mechanisms, communication strategies, and system design considerations for underwater swarm robotics. It examines key marine-specific algorithms, including the Artificial Fish Swarm Algorithm, Whale Optimisation Algorithm, Coral Reef Optimisation, and Marine Predators Algorithm, highlighting their applications in formation control, task allocation, and environmental interaction. The review also analyses communication constraints unique to the underwater domain and emerging acoustic, optical, and hybrid solutions that support cooperative operation. Additionally, it examines hardware and system design advances that enhance system efficiency and scalability. A multi-dimensional classification framework evaluates existing approaches across communication dependency, environmental adaptability, energy efficiency, and swarm scalability. Through this integrated analysis, the review unifies bio-inspired coordination algorithms, communication modalities, and system design approaches. It also identifies converging trends, key challenges, and future research directions for real-world deployment of underwater swarm systems.