Agricultural machinery operates under variable loads, impacts, dust, and changing soil–crop interactions, allowing faults to propagate through energy, material, information, and control pathways. This qualitative review synthesizes 195 research publications across fault formation, trustworthy diagnosis, prognostics and proactive risk control, maintenance and recovery, and case-based evidence assessment. Empirical findings are reported separately from review-derived recommendations, the authors’ conceptual requirements, and mandatory provisions of applicable standards or law. The literature most consistently supports controlled-fault identification, selected single-machine field monitoring, and localized operational compensation. Evidence is weaker for transfer across machines and seasons, calibrated prognostics, safety authorization, post-repair verification, and fleet-scale deployment. On this basis, the review recommends mission profile-specific fault boundaries, traceable diagnostic outputs, explicit uncertainty and abstention, and risk decisions linked to remaining work and resources. It also proposes a diagnostic passport, risk evolution trajectory, repair-effectiveness label, case-evidence matrix, and closed-loop repair workflow as review-level organizing constructs. No included study continuously followed the same machine or fleet through diagnosis, prognosis, authorization, maintenance, acceptance, and feedback; the framework therefore connects evidence-supported stages theoretically rather than claiming end-to-end validation.

Field crops change continuously across growth stages, exhibit substantial spatial heterogeneity, and must be managed within short operational windows. This study presents a structured narrative review of information sensing and intelligent monitoring from pre-sowing conditions to stand establishment, growth and yield formation, biotic stress, maturity, lodging, and harvest readiness. The literature is organized by growth stage and analyzed through a common chain of agricultural need, observable variable, sensing platform, data processing method, validation design, state interpretation, and management or equipment output. Satellite remote sensing, unmanned aerial vehicle sensing, ground and proximal sensing, field Internet of Things, machinery-mounted sensors, multisource fusion, crop models, and machine learning methods are compared according to spatial support, temporal continuity, scale matching, field robustness, transfer conditions, uncertainty, and operational applicability. The reviewed studies report crop-phenotype retrieval, field-environment characterization, and biotic-stress identification under specified conditions, whereas cross-stage state inheritance, consistent reference measurements, independent validation, and conversion of monitoring results into executable tasks remain insufficiently established. The review therefore develops a lifecycle-oriented information-processing perspective in which multisource observations are quality-marked, interpreted as stage states, linked across time and scale, and checked against management and equipment records. Future work should strengthen cross-crop and cross-region validation, mechanistic and data-driven model coordination, uncertainty reporting, interoperability, and field feedback without presuming universally autonomous decision-making.

Post-harvest sorting is essential for converting the biological variability of chili peppers into consistent commercial grades, efficient processing, and higher market value. Rapid advances in machine vision, multimodal sensing, deep learning, and intelligent actuation are transforming sorting equipment from rule-based classifiers into integrated perception-to-execution systems. However, existing studies often evaluate isolated algorithms or components, leaving limited evidence that recognition accuracy translates into reliable sorting during continuous operation. We conducted a structured narrative review of English-language studies published from January 2008 to July 2026 using Web of Science, Scopus, AESC, and PubMed. Evidence was synthesized along a perception-to-execution chain, distinguishing pepper-specific sorting systems and component studies from cross-crop engineering evidence. Direct system-level evidence was concentrated in a small number of chili and bell pepper studies, whereas much of the technical discussion drew on apple, tomato, potato, sweet potato, and onion research. Visible, spectral, fluorescence, and multimodal features can distinguish ripeness, color grades, and surface defects, although performance remains sensitive to cultivar, illumination, pose, and dataset design. A small number of pepper sorting studies support the feasibility of integrating vision, conveying, and physical separation under specific operating conditions. Cross-crop studies inform the discussion of localization, pneumatic actuation, and system integration, but do not establish the performance of these approaches in chili pepper sorting. Reported performance is difficult to compare because studies rarely standardize latency, target association, false and missed rejections, product damage, energy use, and long-term reliability. Robust deployment therefore requires cross-batch datasets, synchronized target-level traceability, and evaluation protocols linking perception outputs to final physical destinations. This review identifies the boundaries of current pepper-specific evidence and proposes a system-level evaluation framework and research priorities for testing the transferability of cross-crop engineering approaches to pepper sorting.

Inland waters such as lakes, reservoirs, rivers, fish ponds, and confined hydraulic structures impose turbidity, shallow-water acoustic multipath, dense boundaries, dynamic biological interference, and limited communication on underwater robots. These coupled constraints can simultaneously degrade sensing, localization, mapping, planning, and control. This structured narrative review synthesizes peer-reviewed English-language evidence on underwater navigation, with emphasis on multisensor fusion localization and the transfer of localization uncertainty into risk-aware planning. Visual, acoustic, inertial, velocity, depth, and external positioning measurements are compared by complementarity, observability, degradation modes, and integration cost. Filtering, sliding-window optimization, factor graphs, estimator switching, and hybrid model- and data-driven approaches are evaluated according to accuracy, real-time performance, robustness, and localization credibility. The review examines how covariance, sensing quality, collision probability, energy, platform dynamics, and task requirements affect maps, path costs, safety margins, trajectory tracking, and degraded operation. Evidence from simulation, public datasets, hardware-in-the-loop tests, tanks, and real waters is synthesized conditionally because study platforms, environments, ground truth, and metrics are heterogeneous. Reliable inland-water autonomy requires diagnosable heterogeneous sensing, integrity-aware localization, and coordinated feedback among localization, planning, and control. Mission-level evaluation should consider data validity, fault recovery, and safe completion rather than average localization error or shortest path alone.

Agricultural weeding robots are progressing from single-function automation toward end-to-end autonomy that integrates task planning, field navigation, weed perception, precision execution, and operational support. This review synthesises 217 sources identified through structured searches of major multidisciplinary, engineering, and Chinese literature databases, supplemented by citation tracking, followed by documented screening, evidence-use coding, technical-theme mapping, and a review-defined L1–L5 validation-coverage assessment. Three main findings emerge. First, reported technical evidence is concentrated at physical closed-loop feasibility in real agricultural environments rather than at complete production service. Among 157 studies with classifiable technical validation, 78 (49.7%) attained L3 as their highest level, whereas only one met the predefined criteria for complete production service and sustained operation with quantified supervision; that study concerned area-based mowing rather than target-level selective weeding. A separate analysis of 49 direct robotic-weeding technical studies showed the same concentration at L3. In a scenario-based boundary-robustness analysis, the number of direct studies attaining L4 or above increased from 1/49 under the strict criteria to 2/49 and 3/49 under two deliberately relaxed scenarios, while L3 remained the dominant highest level. Second, validation depth is uneven: navigation, trajectory control, and several execution modalities have stronger real-field evidence than global task planning, weed perception, fault diagnosis, and autonomous replenishment. Third, the principal system bottleneck is increasingly cross-module rather than purely algorithmic: spatial, temporal, and task-state inconsistencies can propagate through interfaces into missed or repeated treatment, crop damage, unfinished tasks, additional downtime, and human intervention. The review therefore argues for validation centred on recoverable complete production tasks, with explicit reporting of task completion, agronomic outcomes, faults and recovery, replenishment, downtime, resource use, and supervision.

Sweet pepper is a high-value crop in protected horticulture whose robotic harvesting is complicated by canopy occlusion, asynchronous ripening, the need for precise peduncle cutting, and stringent damage limits. This review presents a systematic search and narrative synthesis guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement, with the revised evidence base comprising 262 eligible reports/publications. Reports describing the same robot platform, trial series, or overlapping evaluation dataset were linked as companion reports within a common study/evaluation family to avoid double counting in system-level comparisons. It examines crop constraints, perception, decision-making and system integration, manipulators and end-effectors, and mobile platforms across the complete harvesting cycle. Beyond a module-based synthesis, we propose a five-stage chain—visibility, assessability, reachability, operability, and sustained operation—supported by a unified failure taxonomy, a stage-wise funnel, and a scenario–technology matching matrix. This framework explains why strong module-level results do not necessarily yield effective whole-system harvesting. Representative system-level evaluations within the direct sweet-pepper/Capsicum evidence base report harvesting success rates of approximately 18–83% and source-reported cycle times of approximately 10–69 s per fruit under differing task boundaries, although direct ranking is inappropriate because environments, crop modifications, occlusion levels, sample sizes, task boundaries, denominators, and success criteria differ. As fruit detection and basic grasping improve, losses increasingly arise from incomplete action-ready information on maturity, pose, and peduncles; narrow approach and cutting tolerances; and weak recovery, unloading, and repositioning loops. Near-term development should investigate robot-friendly greenhouses, repeatable docking, local closed-loop control, and human-supervised or collaborative operation, while directly validating productivity, safety, workload, and economic effects. Medium-term priorities are active multimodal perception, fault-tolerant grasping and cutting, and coordinated mobile manipulation. Longer-term progress will depend on agronomy–equipment co-design and long-duration validation of reliability and economic viability.