In recent years, the advent of precision agriculture has accelerated the development of agricultural drone technology. Compared to manual operations, agricultural drones offer higher efficiency, leading to widespread adoption. However, current research on agricultural drones primarily focuses on technical implementation, with relatively little attention from an industrial design perspective. For instance, a linear rotor layout proposed by research teams effectively addresses issues such as narrow spray width and airflow interference in traditional multi-rotor drones. Yet, these products remain in the functional testing phase, neglecting user needs, product functionality, and aesthetics, which hinders the translation of technology into practical productivity. As products mature, meeting user demands, improving quality and performance, and refining functions will become critical factors determining the competitiveness of agricultural drone products. This study aims to propose an integrated theoretical model combining Analytic Hierarchy Process (AHP), Axiomatic Design (AD), and Fuzzy Comprehensive Evaluation (FEC) to optimize the design of a linear tilt-wing agricultural drone based on user needs. We design a linear tilt-wing agricultural drone that satisfies user requirements and provide theoretical insights for similar design endeavors.
Agricultural drones currently available can be categorized into several types: 1. Agricultural helicopters, which are costly, require high pilot skills, and impose economic pressure on farmers. 2. Fixed-wing agricultural drones, characterized by high flight altitudes and long endurance, but with weak downwash airflow, poor droplet penetration, and the need for professional operators. 3. Electric multi-rotor agricultural drones, which are low-cost and easy to use, making them prevalent in China. However, they suffer from limited payload capacity, narrow spray width, and significant airflow interference between rotors, affecting operational accuracy. To address these issues, a linear rotor layout has been proposed, which increases the spraying area and ensures uniform downwash airflow distribution, thereby enhancing operational effectiveness and efficiency. Nonetheless, this product is still in the laboratory stage, with research emphasis on technical application rather than industrial design considerations, resulting in mismatches between aesthetics, performance, and functionality. Examples include piled-up functional components, exposed electronics, poor overall造型, inadequate aesthetics, unreasonable尺寸design, and complex operational steps. This study utilizes AD as a theoretical foundation to guide the transformation of user needs into design technical parameters, integrates AHP for weighting user needs, and applies FEC to evaluate the resulting设计方案, thereby overcoming the limitations of individual theoretical methods and producing a product design theoretical model from a user-need perspective.
The theoretical framework of this research is built upon three core methodologies. Axiomatic Design (AD), established by Professor SUH, aims to create a rigorous axiomatic system in the design field, making the design process more objective and scientific. It includes domains, design constraints, zigzag mapping, the Independence Axiom, and the Information Axiom. The AD design process is driven by user needs through mapping between domains, facilitating the conversion of user needs into specific design parameters and clarifying design goals. The mapping process in AD involves transforming the customer domain (CA) into the functional domain (FR), and then into the physical domain (DP). The relationship is expressed as:
$$ FR = A \times DP $$
where A is the design matrix. The Independence Axiom requires that the design matrix be diagonal or triangular to ensure that functional requirements are independent. Analytic Hierarchy Process (AHP), developed by Saaty, is a multi-criteria decision-making method that reflects subjective judgments. It decomposes goals into hierarchies, compares multiple elements pairwise, and uses weight integration for decision-making, categorizing needs into goal, criterion, and sub-criterion layers. AHP is often used to quantify user needs in design research. However, while AHP can weight user needs, it cannot convert them into design parameters. Fuzzy Comprehensive Evaluation (FEC), introduced by L.A. Zadeh, applies fuzzy mathematics to transform qualitative evaluations into quantitative assessments based on membership degrees. It effectively handles uncertainties and is commonly combined with AHP in design evaluations. Despite its utility, existing research on agricultural drone design using FEC is scarce and often limited to外观造型, without addressing practical constraints such as structural limitations, material rationality, or spray range effects.
To address these gaps, we propose an integrated theoretical model that combines AD, AHP, and FEC. This model leverages the strengths of each method: AHP for weighting user needs, AD for translating needs into design parameters, and FEC for comprehensive evaluation of design solutions. The model’s derivation process involves: (1) acquiring user needs and design constraints for the target product, and integrating and classifying them; (2) applying AHP to weight user needs; (3) using AD theory, under design constraints, to transform user needs into design parameters; (4) conducting design practice based on design parameters; and (5) employing FEC to comprehensively evaluate the resulting方案. This integrated approach ensures a systematic design process for the linear tilt-wing agricultural drone, focusing on user needs and practical applicability.
To deeply understand the current usage, issues, and user needs of agricultural drones, we conducted field research at service centers and research teams. Using qualitative methods such as observation and interviews, we engaged with agricultural drone users, maintenance personnel, sales staff, and technical developers. The results were integrated and classified using the Affinity Diagram (KJ) method, yielding four major categories: functional needs, aesthetic needs, safety needs, and performance needs. These serve as criterion layers (B1-B4), with 19 specific needs as sub-criterion layers (C1-C19), establishing a hierarchical analysis model and design constraints for the linear tilt-wing agricultural drone. The design constraints include technical specifications, regulatory standards, environmental factors, and cost limitations, which must be adhered to during the design process to ensure feasibility and compliance.
| Criterion Layer | Weight | Sub-criterion Layer | Weight | Relative Weight | Rank |
|---|---|---|---|---|---|
| Functional Needs (B1) | 0.2390 | Foldable Main Arm (C1) | 0.3602 | 0.0861 | 4 |
| Easy Transport (C2) | 0.0948 | 0.0227 | 10 | ||
| Easy Battery Installation (C3) | 0.2261 | 0.0540 | 7 | ||
| Night Lighting (C4) | 0.0305 | 0.0113 | 15 | ||
| Detachable Tank (C5) | 0.0475 | 0.0073 | 17 | ||
| Remaining Liquid Detection (C6) | 0.0739 | 0.0177 | 13 | ||
| Visual Obstacle Avoidance (C7) | 0.1670 | 0.0399 | 8 | ||
| Aesthetic Needs (B2) | 0.0529 | Aerodynamic Conformity (C8) | 0.2618 | 0.0139 | 14 |
| Modular Design (C9) | 0.4162 | 0.0220 | 11 | ||
| Symmetrical Shape (C10) | 0.1611 | 0.0085 | 16 | ||
| Aesthetically Pleasing (C11) | 0.1610 | 0.0085 | 18 | ||
| Safety Needs (B3) | 0.5757 | Component Protection (C12) | 0.4658 | 0.2682 | 1 |
| Corrosion Resistance (C13) | 0.0960 | 0.0553 | 6 | ||
| Structural Reliability (C14) | 0.1611 | 0.1596 | 2 | ||
| Waterproofing (C15) | 0.2771 | 0.0927 | 3 | ||
| Performance Needs (B4) | 0.1323 | Lightweight Design (C16) | 0.2973 | 0.0393 | 9 |
| Ergonomic Conformity (C17) | 0.1638 | 0.0713 | 5 | ||
| Lightweight Materials (C18) | 0.5390 | 0.0217 | 12 |
Determining need weights is crucial for objectively evaluating the agricultural drone and guiding subsequent design directions. To ensure scientific rigor, we designed an AHP questionnaire based on the hierarchical model and invited 20 experts, including industrial design professors and practitioners, drone technology researchers, and actual agricultural drone users, to complete it. Using Saaty’s 1-9 scale method, we compared user needs pairwise, assigned scores, and constructed judgment matrices. The geometric mean method was applied to calculate the weights for the agricultural drone user needs, as shown in the table above. Consistency checks were performed to ensure logical合理性; the results are considered reliable only when the consistency ratio (CR) is less than or equal to 0.1. The formulas for consistency check are as follows:
$$ \lambda_{max} = \text{maximum eigenvalue of the judgment matrix} $$
$$ CI = \frac{\lambda_{max} – n}{n – 1} $$
$$ CR = \frac{CI}{RI} $$
where CI is the consistency index, RI is the random index (values depend on n, the matrix order), and CR is the consistency ratio. For all criterion and sub-criterion layers, CR was found to be less than 0.1, confirming data reliability. The weights indicate that safety needs are the most critical, followed by functional needs, performance needs, and aesthetic needs, highlighting priorities for the agricultural drone design.
After weighting user needs, we applied AD theory to translate them into specific design parameters, addressing implementation issues. Based on industry research, user studies, theoretical analysis, and design constraints, we mapped the customer domain (CA) to the functional domain (FR). The mapping results are summarized in the table below:
| Customer Domain (CA) – Need Description | Functional Domain (FR) – Function Description | |
|---|---|---|
| Component Protection | Body Shell Protection | |
| Structural Reliability | Stable Body Platform Structure | |
| Waterproofing | Waterproof Function | |
| Foldable Main Arm | Folding Function | |
| Ergonomic Conformity | Scientific Ergonomic Dimensions | |
| Corrosion Resistance | Anti-corrosion Function | |
| Easy Battery Installation | Detachable Battery Function | |
| Visual Obstacle Avoidance | Visual Obstacle Avoidance Function | |
| Lightweight Design, Lightweight Materials | Lightweight Body Design | |
| Easy Transport | Transport Function | |
| Modular Design | Modular Layout | |
| Remaining Liquid Detection | Remaining Liquid Detection Function | |
| Aerodynamic Conformity | Streamlined Body | |
| Night Lighting | Lighting Function | |
| Symmetrical Shape | Symmetrical Appearance | |
| Detachable Tank | Tank Detachment Function | |
| Aesthetically Pleasing, Design Sense | Design-sense Appearance |
Next, we transformed functional requirements (FR) into design parameters (DP). To ensure validity and科学性, we referenced national standards for agricultural drone safety, along with academic research on agricultural drones and AD applications. For each functional requirement, corresponding design parameters were proposed, and each parameter was checked against design constraints; non-compliant parameters were modified or removed. During mapping, FR2 (stable body platform structure) was decomposed into FR21 (stable arm structure) and FR22 (stable frame structure) for separate mapping. The results are shown in the table below:
| Functional Domain (FR) – Function Description | Physical Domain (DP) – Design Parameter Description |
|---|---|
| Body Shell Protection | ABS Plastic Shell |
| Stable Body Platform Structure | Carbon Fiber + Aluminum Alloy Main Structure |
| Stable Arm Structure | Carbon Fiber Rod (Diameter 300mm) |
| Stable Frame Structure | Aluminum Alloy Frame |
| Waterproof Function | IP67-grade Protective Adhesive |
| Folding Function | Safety Hinge Folding |
| Scientific Ergonomic Dimensions | 2500 mm × 525 mm × 525 mm |
| Anti-corrosion Function | Core Component Anti-corrosion |
| Detachable Battery Function | Battery Plug-in Structure |
| Visual Obstacle Avoidance Function | Binocular Vision Obstacle Avoidance Sensor |
| Lightweight Body Design | Lightweight Material Application |
| Transport Function | Single-person Carrying Handle |
| Modular Layout | Component and Appearance Modularization |
| Remaining Liquid Detection Function | Liquid Level Sensor Application |
| Streamlined Body | Streamlined Shape and Fade Surfaces |
| Lighting Function | 8W, 120° LED Lighting |
| Symmetrical Appearance | Symmetrical Appearance Design |
| Tank Detachment Function | Detachable Tank Structure Design |
| Design-sense Appearance | Aesthetic and Design-sense造型 |
According to the Independence Axiom, the design parameters must be checked. The relationship between design parameters and functional requirements is expressed as:
$$ FR_i = \sum_{j=1}^{n} A_{ij} DP_j \quad \text{for} \quad i=1,2,\ldots,m $$
where A is the design matrix for the linear tilt-wing agricultural drone, and A_{ij} represents the correlation between functional requirement FR_i and design parameter DP_j:
$$ A_{ij} = \frac{\partial FR_i}{\partial DP_j} $$
In this equation, m is the number of functional requirements (FR), and n is the number of design parameters (DP). The design matrix A can take three forms: diagonal (uncoupled), triangular (decoupled), or general (coupled). In AD theory, “0” indicates no correlation, and “1” indicates correlation. A diagonal matrix satisfies the Independence Axiom, allowing design parameters to be used; a triangular matrix requires reordering; a general matrix violates the axiom. Based on the mapping, we derived the design matrix A for the linear tilt-wing agricultural drone:
$$ A = \begin{bmatrix}
1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1 & 0 \\
0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 1
\end{bmatrix} $$
Comparing with matrix forms, this design matrix is diagonal, satisfying the Independence Axiom. Thus, design parameters DP1 to DP17 are feasible and can serve as the basis for design practice.
Based on the AD parameter mapping results, we conducted innovative design for the linear tilt-wing agricultural drone. The overall product design features are described below. To ensure robust structural strength while protecting electrical systems (waterproof, dustproof, corrosion-resistant) and achieving lightweight goals, the外壳and main body use ABS and carbon fiber materials, with an aluminum alloy frame. IP67-grade protective adhesive is applied to circuit boards. For compact storage and easy user transport, the arms are designed with hinges and carrying handles, with dimensions set at 2500 mm × 525 mm × 525 mm. Given frequent battery changes during agricultural drone operations, a quick-plug battery installation design is implemented. Additionally, a non-contact liquid level sensor is incorporated to meet the need for monitoring remaining liquid. To facilitate maintenance and operation, the appearance and internal structure follow modular and symmetrical principles. To reduce aerodynamic drag during flight, the body外壳is streamlined with fade surfaces. For safety, a visual obstacle avoidance module is added to complement radar avoidance, and lighting functions are included for night use. The integrated design ensures that this agricultural drone meets user needs while maintaining high performance and aesthetics.
To validate the设计方案effectiveness, we employed Fuzzy Comprehensive Evaluation (FCE). First, a comment set V was constructed参照李克特量表: V = {Very Poor, Poor, Average, Good, Excellent}, corresponding to scores of 50, 60, 70, 80, 90. Ten participants from earlier research phases and industrial design experts were invited to rate the设计方案based on sub-criterion layers, resulting in fuzzy evaluation matrices for functional needs (U1), aesthetic needs (U2), safety needs (U3), and performance needs (U4). The matrices are as follows:
$$ U1 = \begin{bmatrix}
0.4 & 0.4 & 0.1 & 0.1 & 0 \\
0.5 & 0.1 & 0.1 & 0.2 & 0.1 \\
0.5 & 0.3 & 0.1 & 0.1 & 0 \\
0.5 & 0.2 & 0.1 & 0.1 & 0.1 \\
0.3 & 0.4 & 0.1 & 0.1 & 0.1 \\
0.1 & 0.7 & 0.1 & 0.1 & 0 \\
0.6 & 0.2 & 0.1 & 0.1 & 0
\end{bmatrix} $$
$$ U2 = \begin{bmatrix}
0.7 & 0.2 & 0 & 0.1 & 0 \\
0.7 & 0.3 & 0 & 0 & 0 \\
0.4 & 0.5 & 0.1 & 0 & 0 \\
0.7 & 0.3 & 0 & 0 & 0
\end{bmatrix} $$
$$ U3 = \begin{bmatrix}
0.5 & 0.2 & 0.1 & 0.1 & 0.1 \\
0.7 & 0.2 & 0.1 & 0 & 0 \\
0.6 & 0.2 & 0.1 & 0.1 & 0 \\
0.7 & 0.2 & 0.1 & 0 & 0
\end{bmatrix} $$
$$ U4 = \begin{bmatrix}
0.8 & 0.1 & 0.1 & 0 & 0 \\
0.7 & 0.1 & 0.1 & 0.1 & 0 \\
0.6 & 0.2 & 0.1 & 0.1 & 0
\end{bmatrix} $$
From Table 1, the criterion layer weight vector is P_B = (0.2390, 0.0529, 0.5757, 0.1323). Similarly, the sub-criterion layer weight vectors for functional needs (P_B1), aesthetic needs (P_B2), safety needs (P_B3), and performance needs (P_B4) were calculated. The evaluation weight matrix W for the sub-criterion layers is derived:
$$ W = \begin{bmatrix}
0.44162 & 0.33162 & 0.1 & 0.10948 & 0.01728 \\
0.65174 & 0.30607 & 0.01611 & 0.02618 & 0 \\
0.59073 & 0.2 & 0.1 & 0.06269 & 0.04658 \\
0.6758 & 0.15391 & 0.10001 & 0.07028 & 0
\end{bmatrix} $$
Using the criterion layer weight vector P_B and the evaluation matrix W, the comprehensive evaluation vector P is computed:
$$ P = P_B \times W = (0.56952, 0.23095, 0.09555, 0.07294, 0.03095) $$
Then, the final score is obtained by multiplying P with the score vector S = (50, 60, 70, 80, 90):
$$ \text{Final Score} = 0.56952 \times 50 + 0.23095 \times 60 + 0.09555 \times 70 + 0.07294 \times 80 + 0.03095 \times 90 = 82.3452 $$
This score falls within the “Good” range, indicating that the设计方案largely satisfies user needs and that the integrated AHP-AD-FEC theoretical model effectively guides the optimization design of the linear tilt-wing agricultural drone.
With accelerating urbanization, rural labor reduction, and land intensification policies, agricultural mechanization is inevitable. Drone technology plays a key role in advancing this trend. In optimizing the linear tilt-wing agricultural drone design, we used AD as a framework to translate user needs into technical parameters, clarifying design goals. AHP was applied to weight user needs, and FEC evaluated the设计方案. The results show that this approach effectively enhances user satisfaction,验证了the feasibility of the integrated AHP-AD-FEC model, and further refines the product design process. This study not only contributes to the development of agricultural drones but also offers a systematic methodology for similar product designs, emphasizing the importance of user-centric approaches in technological innovation. Future work could explore real-world testing of the designed agricultural drone to validate its performance in field conditions and further refine the model based on feedback.