Design and Evaluation of Shared Tennis Service Robots Based on AHP–FCE

Abstract

To address persistent challenges in tennis—such as inefficient ball retrieval, the high cost of serving equipment, and difficulties in scheduling matches—this study proposes the design of a shared tennis service robot aimed at improving user experience and validating design feasibility. Grounded in user experience theory, user requirements were collected through questionnaires and structured interviews. The Analytic Hierarchy Process (AHP) was adopted to construct a hierarchical model of requirements. Weighted calculations were then applied to quantify and rank user needs. Design solutions were then derived based on these rankings. To evaluate the solutions, the Fuzzy Comprehensive Evaluation (FCE) method was utilized for multidimensional assessment. The results show that AHP identified three core requirements: intelligent ball retrieval, intelligent serving, and personalized serving parameter customization. Guided by these priorities, the proposed design integrates a shared rental model with multisensory interactive feedback. The final evaluation yielded an FCE score of 87.83, confirming the effectiveness of the solution. The combined AHP-FCE method provides a systematic framework for quantifying user needs and objectively evaluating design alternatives. It also offers a methodological foundation for the development of sports service robots. The shared tennis robot effectively reduces labor and operational costs while enhancing the overall user experience.

1. Introduction

With the increasing popularity of tennis, more enthusiasts are engaging in the sport [1,2]. However, during training sessions, large numbers of tennis balls are used and often scattered across the court, making retrieval both time-consuming and physically demanding [3]. In professional competitions or elite training, this task is typically handled by ball boys or girls, but most recreational players must collect balls manually. Although commercial ball retrievers are available, they are primarily mechanical or manual devices with limited intelligence. They also lack advanced features such as automated navigation or centralized collection. On larger courts, players still expend considerable effort in pushing or maneuvering equipment to complete ball retrieval, which diminishes the overall tennis experience [4,5].

The rapid expansion of the modern market economy has firmly established the internet as a driving force of contemporary economic development [6]. The rise of social networking platforms, along with advances in smart devices and logistics systems, has accelerated the growth of the sharing economy [7,8]. By leveraging online platforms, the sharing economy facilitates more efficient resource allocation, thereby enhancing utilization efficiency and improving public convenience. From the consumer perspective, the ideal consumption model is shifting from product ownership to temporary access, reflecting a growing preference for short-term use of goods or services rather than long-term possession [9,10]. In the context of tennis, although the sport is gaining popularity, it remains relatively niche in China compared to other sports. As a result, players often face difficulties in finding partners for matches, while the high cost of ball machines further restricts participation [11,12].

In recent years, researchers have increasingly focused on the intelligent development of tennis auxiliary equipment and innovations in service models. Wang et al. introduced a path-planning technique for intelligent ball-retrieval robots that integrates a twin-network target tracking algorithm, improving ball-picking efficiency and obstacle avoidance [4]. However, their work was limited to the single function of ball retrieval. Similarly, Wu and Xiao enhanced deep learning algorithms in order to enhance the precision of tennis ball recognition [13]. Their work demonstrated the feasibility of visual recognition for ball detection, but it did not address multifunctional design or user experience. Within the sharing economy domain, Kim and Lee examined how the features of sports-sharing services affect usage intention, identifying convenience, trust, and maintenance mechanisms as critical success factors [14]. Nevertheless, existing research has not yet provided a systematic evaluation of user experience that integrates ball retrieval, ball serving, and shared rental functions. This gap defines the entry point for the present study.

Therefore, integrating the concept of the sharing economy with research on tennis service robots provides an opportunity to design a shared service robot specifically for tennis. This approach aligns with user needs, allowing more players to benefit from the convenience of the sharing economy while enhancing the overall sporting experience. The sharing economy model introduces new strategies for reducing usage costs, whereas advances in intelligent service robots provide the technological foundation for functional integration. Accordingly, this study proposes the design of a tennis service robot that combines ball retrieval, ball feeding, and shared rental functions. Grounded in user experience theory and employing the AHP-FCE method, the proposed robot seeks to address the dual challenges of resource underutilization and suboptimal athletic experiences.

2. Methods

2.1. Solution Approach and Process

This study adopts a multi-criteria decision analysis framework. Following a comparative review of available methodologies, the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE) were selected as the core tools, establishing a closed-loop design process that extends from demand quantification to solution validation.

The Analytic Hierarchy Process (AHP), developed by Professor Thomas L. Saaty at the University of Pittsburgh, is a systematic and hierarchical analytical method that integrates qualitative and quantitative approaches [15]. Its core principle is to decompose factors related to the target problem into multiple levels—such as objectives, criteria, and alternatives—thereby constructing a top-down hierarchical model [16]. Within this structure, judgment matrices are created for each element, and relative weights are calculated to determine the importance ranking of factors, facilitating the identification of core elements [17,18]. The strength of AHP lies in its ability to analyze the essence, key determinants, and interrelationships of complex evaluation problems. At the same time, it translates qualitative judgments into numerical values, thereby streamlining and objectifying the assessment process.

Fuzzy Comprehensive Evaluation (FCE) is a method grounded in fuzzy mathematics that applies fuzzy relational synthesis theory to address factors with indistinct boundaries and characteristics that are difficult to quantify [19,20]. The approach adopts a hierarchical structure, decomposing complex systems step by step into target, criterion, and indicator layers, while integrating subjective and objective weights to enable multidimensional quantitative analysis [21,22]. FCE constructs fuzzy subsets for evaluation levels to determine the membership degrees of indicators representing the evaluation object. These indicators are then synthesized through fuzzy transformation principles, producing a holistic evaluation within a comprehensive analytical framework [23].

In this study, the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE) are integrated to establish a closed-loop design research framework. AHP is employed to hierarchically model user requirements and perform quantitative calculations, thereby identifying core design priorities. FCE is then applied to objectively evaluate proposed design solutions across multiple dimensions, ensuring that the final outcome aligns with user needs and remains logically consistent [24,25]. Incorporating FCE creates a complete loop that spans requirement quantification, design generation, and outcome validation. This integration effectively addresses AHP’s limitation in evaluating whether design results align with user expectations [25,26,27]. This integration ensures both the scientific rigor of requirement analysis and the objectivity of solution evaluation.

This study adopts a user experience design perspective, leveraging the sharing economy model as an entry point to investigate and analyze user needs. A service process and architectural framework for shared tennis service robots is constructed, providing a theoretical foundation for subsequent research. To further define the target audience, the Analytic Hierarchy Process (AHP) is applied to both qualitative and quantitative studies of tennis activity requirements, enabling in-depth analysis of user needs. Using AHP, a key design element matrix is developed to calculate demand weights and establish a hierarchy of priorities. These priorities are then translated into concrete design elements, ensuring stronger alignment between product features and user needs. High-weight user requirements are incorporated into product functionalities, which undergo multidimensional positioning analysis before informing the design and interface development of the shared tennis service robot—from sketching to 3D modeling and rendering. Finally, the design is validated through Fuzzy Comprehensive Evaluation (FCE) to confirm its rationality and scientific rigor. The complete research workflow is presented in Figure 1.

2.2. User Requirements Gathering

2.2.1. User Questionnaire Survey

This study utilizes a combination of online and offline questionnaires to collect comprehensive data on users’ personal information, usage behaviors, and habits [28]. A total of 200 questionnaires are planned for distribution, with a target of at least 150 valid responses to ensure an adequate and representative sample size. The survey is primarily directed toward tennis enthusiasts, and the detailed process is illustrated in Figure 2.

This questionnaire survey adopted a semi-structured format. Most items included predefined options to ensure reliability and facilitate data analysis, while a smaller number of open-ended questions were designed to capture more detailed responses [29]. As outlined in

Table 1. Survey Questionnaire Design Outline.

Serial Number	Question Type	Brief Description of the Issue
1	Basic Information	Gender, age, occupation, city of residence, income
2	Behavioral Information Section	Frequency of playing on the court; Practice methods (playing with partners, using solo training aids, coaching sessions); Challenges encountered (difficulty finding playing partners, training aids being too expensive)
3	Product Experience (Serving)	Usage Status—Reasons for Non-Use, Whether Interaction Process Is Cumbersome
4	Product Experience (Ball Retrieval)	Ball retrieval time, ball retrieval method, inconveniences/shortcomings
5	Product Expectations Section	Serve Function Preferences (Adjustable/Fixed), Interaction Methods (Remote Control, Voice, App, Physical Buttons), Personalization Needs, Interaction Feature Requirements (Light Indicators, Remote Control), Esthetic Preferences (Sporty Style…), Willingness to Try Sharing, Rental Cost Range, Key Features Desired?

, the questionnaire consisted of five sections. The first section collected user information and exercise habits to refine target user segmentation. The second focused on user needs during sports activities, with structured questions addressing both exercise experience [30] and product experience [31,32], particularly ball retrieval and serving. The final section examined product expectations, gathering user requirements across three dimensions: functional needs, interaction needs, and esthetic preferences [33].

2.2.2. User Interviews

Following the collection of questionnaire data, additional interviews were conducted with several representative target users, as summarized in

Table 2. User Interview Outline.

Interview Stage	Main Section	Main Issues
Initial stage	Personal Information	Gender, age, education level, hobbies
	Before playing ball	Tennis Activity Status, Practice Schedule, Playing Partners Status, Ball Machine Usage Status
Intermediate stage	Playing ball	Answering and making calls, dropping and picking up items, interaction feedback methods
	After playing ball	Ball Retrieval Methods, Ball Retrieval Duration, Practice Logs, Issues with Ball Retrieval
In-depth Phase	User Expectations	Basic functional requirements, interaction requirements, visual design requirements

. Using a semi-structured format [34], these interviews explored participants’ exercise habits, tennis experiences, and multifaceted needs for shared tennis service robots. The findings provided a foundation for defining design directions and refining both the robot’s design and service features. To avoid bias, leading questions were excluded, and participants were given sufficient time for reflection. When personal insights were shared, the researcher engaged in active listening and offered appropriate feedback.

2.3. AHP Hierarchical Analysis Method Model Construction

2.3.1. Establish a Hierarchical Model

First, a hierarchical structural model must be established, consisting of the objective layer, the criterion layer, and the indicator layer. The objective layer generally represents the ultimate goal to be achieved through the decision-making process.

2.3.2. Construct a Decision Matrix

After constructing the hierarchical model, calculations are carried out to determine the influence of the indicator layer on the criterion layer and the influence of the criterion layer on the objective layer. This process establishes the relative weights within each layer, which are finalized through pairwise comparisons and subsequent computations. When assessing the relative importance of two measures within the identical category, a 1–9 rating system is applied, as summarized in

Table 3. AHP Scaling Method.

Scale Level	1	3	5	7	9	2, 4, 6, 8	1/3, 1/5, 1/7, 1/9
Semantic Evaluation	Equally important	Slightly important	Important	Extremely important	Of paramount importance	The median value in two consecutive scales	The ratio of the importance of indicators $i$ and $j$ is denoted as $a_{ij}$. Then, the importance ratio of indicator $i$ to $j$ satisfies $a_{ij}=1/a_{ij}$

. For example, in the criterion layer, determining the weight of each criterion relative to the objective layer involves comparing n elements, the results of which are used to construct a judgment matrix:

$$C=\left(C_{ij}\right)n\times n=\left[\begin{array}{ccc}{C}_{11}& \cdots & C_{1n}\\ ⋮& \ddots & ⋮\\ C_{n1}& \cdots & C_{nn}\end{array}\right]$$

(1)

2.3.3. Calculate the Weight Value

Normalize each column element in the decision matrix C:

$$a_{ij}={\displaystyle \frac{c_{ij}}{{\sum }_{i=1}^n{c}_{ij}}}$$

(2)

Sum the rows of the normalized decision matrix:

$$\overline{w_i}=\sum _{j=1}^n{a}_{ij}$$

(3)

After weight transformation of the normalized elements yields the weight vector $w^1={\left(w_1^1, w_2^1,\dots , w_n^1\right)}^T$, the subjective weights for each indicator can be obtained:

$$w_i={\displaystyle \frac{\overline{w_i}}{{\sum }_{i=1}^n\overline{w_i}}}$$

(4)

2.3.4. Consistency Test

To perform consistency testing on the results, first determine the largest eigenvalue of the matrix:

$${\lambda }_{max}={\displaystyle \frac{1}{n}}{\sum }_{i=0}^n{\displaystyle \frac{{\left(cw\right)}_i}{w_i}}$$

(5)

In the formula, ${\left(cw\right)}_i$ denotes the $i$-th component of vector $cw$, and then the consistency metric is calculated.

$$CI={\displaystyle \frac{{\lambda }_{max}-n}{n-1}}$$

(6)

Finally, the random consistency index RI value of the judgment matrix is introduced, as shown in

Table 4. Random Coincidence Index $RI$ Values.

degree	3	4	5	6	7	8	9	10
$RI$	0.58	0.89	1.12	1.24	1.32	1.41	1.45	1.49

. When $CR={\displaystyle \frac{CI}{CR}}<0.1$, the judgment matrix is considered consistent and the weights are valid.

2.3.5. Calculate the Importance of User Demand for Each Metric

$f\left(D_i\right)$ represents the importance of user requirements at the criterion level, $f\left(d_i\right)$ denotes the importance of requirements at the indicator level, where its importance value equals the indicator weight $w$ at that level, and $f\left({Dd}_i\right)$ signifies the comprehensive importance of user requirements:

$$f\left({Dd}_i\right)=\mathrm{ }f\left(D_i\right)\times f\left(d_i\right)$$

(7)

2.4. Fuzzy Comprehensive Evaluation Method Calculation Steps

2.4.1. Determining the Set of Factors for the Evaluation Subject

A factor set is a collection of factors that influence the evaluation object, typically expressed as $G=\left\{G_1, G_2,{\dots , G}_m\right\}$, where each element $G$ represents the $i$-th factor influencing the assessment subject. These factors typically demonstrate different levels of vagueness.

2.4.2. Set of Evaluation Comments for the Target

The purpose of the comment set is to provide qualitative descriptions for each indicator, with particular emphasis on experts’ qualitative feedback. By organizing this feedback, an evaluation set for each indicator can be constructed, $V=\left\{V_1, V_2,{\dots , V}_k\right\}$, where $V_k$ denotes the rating level corresponding to each factor.

2.4.3. Determine the Comprehensive Evaluation Matrix

By applying fuzzy evaluation to each factor, the corresponding fuzzy relationships $R$ are established. Domain experts are then invited to score all elements and assess the fuzzy relationship matrix. The finalized fuzzy relationship matrix is expressed as:

$$R=\left[\begin{array}{ccccc}{r}_{11}& r_{12}& r_{13}& \cdots & r_{1n}\\ r_{21}& r_{22}& r_{23}& \cdots & r_{2n}\\ \cdots & \cdots & \cdots & \cdots & \cdots \\ r_{m1}& r_{m2}& r_{m3}& \cdots & r_{mn}\end{array}\right]$$

(8)

2.4.4. Comprehensive Evaluation Calculation

By combining the weights $\omega$ derived from AHP calculations with the fuzzy relationship matrix $R$, the fuzzy comprehensive evaluation model $B$ is constructed:

$$B=\omega \times R=\left({\omega }_1, {\omega }_2,\dots , {\omega }_n\right)\times \left[\begin{array}{ccccc}{r}_{11}& r_{12}& r_{13}& \cdots & r_{1n}\\ r_{21}& r_{22}& r_{23}& \cdots & r_{2n}\\ \cdots & \cdots & \cdots & \cdots & \cdots \\ r_{m1}& r_{m2}& r_{m3}& \cdots & r_{mn}\end{array}\right]$$

(9)

In this formula, $B$ represents the fuzzy evaluation vector; $\omega$ denotes the weight set; and $R$ is the fuzzy relationship matrix derived from the evaluation factor set $G$ and the evaluation set $V$.

By converting each indicator into corresponding numerical scores, evaluation results can be obtained, enabling the prioritization of design proposals. If the evaluation set is defined as $V=\left\{V_1, V_2, {V_3, V}_4\right\}$, scores of 90, 80, 70, and 60 are assigned to $V_1, V_2, V_3 \mathrm{a}\mathrm{n}\mathrm{d} V_4$, respectively. Let $Y={\left(90, 80, 70, 60\right)}^T$; the comprehensive decision value is then calculated as follows:

$$Z:Z=B\circ Y$$

(10)

The symbol “$\circ$” in the equation denotes a fuzzy synthesis operation, with common types including the weighted average method and the dominant factor method. In this study, the weighted average operator model $M\left(·,+\right)$ is employed to compute $Z$.

3. Results: Shared Tennis Service Robot Prototype Design

3.1. Survey Results on User Needs for Shared Tennis Service Robots

3.1.1. User Requirements Integration and Extraction

A positive user experience enhances both user engagement and loyalty to a product. This research extracted and condensed user requirements by observing, conducting online research, literature review, questionnaire surveys, and interviews. User research served as the primary method, with target user segments refined by collecting demographic information and behavioral patterns related to sports activities. Particular attention was given to users’ exercise and product usage experiences, especially in the contexts of ball retrieval and ball serving. Structured questions were employed to investigate user requirements for these activities. Finally, product expectations were examined across three dimensions: functional requirements, interaction requirements, and esthetic requirements. Due to uncontrollable factors such as participant attrition during survey collection, 161 questionnaires were gathered. After screening, 156 were deemed valid. Analysis of these valid responses, combined with in-depth interviews with target users, identified 17 secondary demand indicators, which were further categorized into explicit and implicit needs, as shown in Figure 3.

3.1.2. Hierarchical Classification of User Needs

After consolidating and analyzing user requirements, it is essential to systematically categorize and integrate all functional demands identified by users. This process ensures the clarity, coherence, and feasibility of user needs, thereby establishing a robust data foundation and quantitative basis for subsequent Analytic Hierarchy Process (AHP) analysis.

Drawing on Maslow’s Hierarchy of Needs theory, the consolidated user requirements were organized and categorized into four levels of demand for shared tennis service robots: functional needs, sensory needs, interactive needs, and self-actualization needs. The detailed classification is presented in

Table 5. Hierarchical Division of User Needs.

Primary User Requirements	Secondary User Requirements	Tier 3 User Requirements
User Requirements for Shared Tennis Service Robots	Functional Requirements	Smart Ball Retrieval Smart Serve Shared Rental Model Mobility Tennis Storage
	Esthetic needs	Tech-inspired LED light strips Sleek design Everyday color palette
	Interaction Requirements	Clean interface Data visualization User-friendly operation Light indicators Voice interaction
	Self-actualization needs	Customized serving parameters Personalized training plans Community exchange platform

.

3.2. User Requirement Weight Analysis Based on the AHP Method

3.2.1. Constructing an AHP Hierarchical Model

According to the user requirement hierarchy classification (Table 5), a total of 17 requirements were identified. These were further structured into a hierarchical model for the shared tennis service robot using the Analytic Hierarchy Process (AHP). The overall objective layer (E) represents the design plan for the shared tennis service robot, while the criterion layer (P) includes functional, esthetic, interaction, and Self-actualization needs. The subcriterion layer (G) comprises specific requirements such as Smart Ball Retrieval, Smart Serve, and Clean interface. The full hierarchical structure is presented in Figure 4.

3.2.2. Constructing an Evaluation Indicator Judgment Matrix

Following the procedural steps of the Analytic Hierarchy Process (AHP), a judgment matrix questionnaire for the shared tennis service robot was developed. An expert scoring method was employed to determine indicator ratings, thereby enhancing the objectivity and generalizability of the evaluation data. A panel of eight relevant experts was convened, with the following criteria for their selection:

• At least 3 years of relevant professional experience in industry or research;
• Demonstrated practical background and specialized knowledge in service robotics, product design, or tennis sports;
• Possess independent judgment capabilities and be thoroughly familiar with the fundamental processes and application scenarios of tennis sports.

Specifically, the expert panel comprises two senior designers specializing in service robot design, two tennis coaches, two industrial design faculty members, and two master’s students in industrial design engineering. This composition aims to balance technical expertise, sports application, and diverse design perspectives from the user’s standpoint.

Given that the relatively small size of the expert panel may lead to individual opinions exerting a significant influence on the overall results, this study implemented the following measures to minimize bias and enhance the consistency of findings:

• Employ pairwise comparisons to score the importance of indicators at the same level, quantifying them using the Saaty 1–9 scale;
• Calculate the geometric mean of all collected scores to mitigate the impact of individual experts’ extreme opinions on the final judgment matrix;
• Conducted consistency testing (CR) on the resulting judgment matrix to ensure it meets the consistency criterion CR ≤ 0.1. If the criterion was not met, expert opinions were revisited and adjusted;
• Discussed and documented potential expert opinion biases during the results analysis phase to enhance the transparency and reproducibility of the study.

Following the aforementioned process, the criterion layer and sub-criterion layer indicator judgment matrices were constructed based on Formula (1) (see

Table 6. Guideline Layer Decision Matrix E1.

E1 Judgment Matrix	P1 Functional Requirements	P2 Esthetic Needs	P3 Interaction Requirements	P4 Self-Actualization Needs
P1 Functional Requirements	1	9	5	3
P2 Esthetic needs	1/9	1	1/6	1/4
P3 Interaction Requirements	1/5	6	1	1
P4 Self-actualization needs	1/3	4	1	1

,

Table 7. Sub-criterion Level Judgment Matrix P1.

P1 Judgment Matrix	G1 Smart Ball Retrieval	G2 Smart Serve	G3 Shared Rental Model	G4 Mobility	G5 Tennis Storage
G1 Smart Ball Retrieval	1	1	5	7	8
G2 Smart Serve	1	1	6	5	6
G3 Shared Rental Model	1/5	1/6	1	3	2
G4 Mobility	1/7	1/5	1/3	1	4
G5 Tennis Storage	1/8	1/6	1/2	1/4	1

,

Table 8. Sub-criterion Level Judgment Matrix P2.

P2 Judgment Matrix	G6 Tech-Inspired	G7 LED Light Strips	G8 Sleek Design	G9 Everyday Color Palette
G6 Tech-inspired	1	8	1	7
G7 LED light strips	1/8	1	1/6	3
G8 Sleek design	1	6	1	6
G9 Everyday color palette	1/7	1/3	1/6	1

,

Table 9. Sub-criterion Level Judgment Matrix P3.

P3 Judgment Matrix	G10 Clean Interface	G11 Data Visualization	G12 User-Friendly Operation	G13 Light Indicators	G14 Voice Interaction
G10 Clean interface	1	1	4	3	7
G11 Data visualization	1	1	1/2	5	6
G12 User-friendly operation	1/4	2	1	3	4
G13 Light indicators	1/3	1/5	1/3	1	2
G14 Voice interaction	1/7	1/6	1/4	1/2	1

and

Table 10. Sub-criterion Level Judgment Matrix P4.

P4 Judgment Matrix	G15 Clean Interface	G16 Data Visualization	G17 User-Friendly Operation
G15 Clean interface	1	4	7
G16 Data visualization	1/4	1	4
G17 User-friendly operation	1/7	1/4	1

), providing reliable foundational data support for subsequent weight calculations and fuzzy comprehensive evaluation.

3.2.3. Calculate Demand Weighting and Consistency Check

Due to the large number of calculation results, it is not feasible to list them all individually. Here, we present the calculation steps using the criterion-level judgment matrix $E_1$ as a representative example:

1. According to Table 3, from Equation (1), we can determine that the decision matrix $E_1$ is:

$$E_1=\left[\begin{array}{cccc}1& 9& 5& 3\\ 0.1111& 1& 0.1667& 0.25\\ 0.2& 6& 1& 1\\ 0.3333& 4& 1& 1\end{array}\right]$$

where the matrix dimension is n = 4;

2. The weight values for each indicator were calculated separately for both the criterion layer and the indicator layer. The weight vector values were derived using Formula (2). Taking the criterion layer judgment matrix $E_1$ as an example, column normalization was first performed to obtain the normalized matrix $E_2$, as shown below:

$$E_2=\left[\begin{array}{cccc}0.6081& 0.4500& 0.6977& 0.5714\\ 0.0676& 0.0500& 0.0233& 0.0476\\ 0.1216& 0.3000& 0.1395& 0.1905\\ 0.2027& 0.2000& 0.1395& 0.1905\end{array}\right]$$

3. The row-normalized sum of the computational matrix $E_2$ is obtained from Formula (3):

$$\overline{w_1}=0.6081+0.4500+0.6977+0.5714=2.3272$$

$$\overline{w_2}=0.0676+0.0500+0.0233+0.0476=0.1884$$

$$\overline{w_3}=0.1216+0.3000+0.1395+0.1905=0.7516$$

$$\overline{w_4}=0.2027+0.2000+0.1395+0.1905=0.7327$$

The geometric mean of each row of the matrix was calculated, and the resulting maximum eigenvector (weight) was normalized to yield the following:

$$w_i=\left[\begin{array}{c}0.5818\\ 0.0471\\ 0.1879\\ 0.1832\end{array}\right]$$

4. Perform consistency testing on the results. The maximum eigenvalue ${\lambda }_{max}=4.1272$ is calculated from the matrix using Formula (5). Then, according to Formula (6), we obtain:

$$CI={\displaystyle \frac{{\lambda }_{max}-n}{n-1}}={\displaystyle \frac{4.1272-4}{4-1}}=0.0424$$

$$CR={\displaystyle \frac{CI}{RI}}={\displaystyle \frac{0.0424}{0.89}}=0.0476$$

where $CR<0.1$. Then, the decision matrix is consistent, and the weights are valid.

Following the aforementioned steps, a consistency test was conducted for the sub-criteria layer judgment matrix by calculating the indicator weights ($w$), the maximum eigenvalue of the matrix (${\lambda }_{max}$), the consistency index ($CI$), and the consistency ratio ($CR$). The results of these calculations are presented in

Table 11. Sub-criterion Level P1 Evaluation Indicators.

P1 Judgment Matrix	G1 Smart Ball Retrieval	G2 Smart Serve	G3 Shared Rental Model	G4 Mobility	G5 Tennis Storage	Normalized Weight Index
G1 Smart Ball Retrieval	1.0000	1.0000	5.0000	7.0000	8.0000	0.4003
G2 Smart Serve	1.0000	1.0000	6.0000	5.0000	6.0000	0.3722
G3 Shared Rental Model	0.2000	0.1667	1.0000	3.0000	2.0000	0.1009
G4 Mobility	0.1429	0.2000	0.3333	1.0000	4.0000	0.0830
G5 Tennis Storage	0.1250	0.1667	0.5000	0.2500	1.0000	0.0437
Inspection Criteria	${\lambda }_{max}=5.4138\hspace{1em}\hspace{1em}\hspace{1em}CI=0.1035\hspace{1em}\hspace{1em}\hspace{1em}CR=0.0924$

,

Table 12. Sub-criterion Level P2 Evaluation Indicators.

P2 Judgment Matrix	G6 Tech-Inspired	G7 LED Light Strips	G8 Sleek Design	G9 Everyday Color Palette	Normalized Weight Index
G6 Tech-inspired	1.0000	8.0000	1.0000	7.0000	0.4508
G7 LED light strips	0.1250	1.0000	0.1667	3.0000	0.0921
G8 Sleek design	1.0000	6.0000	1.0000	6.0000	0.4034
G9 Everyday color palette	0.1429	0.3333	0.1667	1.0000	0.0537
Inspection Criteria	${\lambda }_{max}=4.1908\hspace{1em}\hspace{1em}\hspace{1em}CI=0.0636\hspace{1em}\hspace{1em}\hspace{1em}CR=0.0715$

,

Table 13. Sub-criterion Level P3 Evaluation Indicators.

P3 Judgment Matrix	G10 Clean Interface	G11 Data Visualization	G12 User-Friendly Operation	G13 Light Indicators	G14 Voice Interaction	Normalized Weight Index
G10 Clean interface	1.0000	1.0000	4.0000	3.0000	7.0000	0.3687
G11 Data visualization	1.0000	1.0000	0.5000	5.0000	6.0000	0.2756
G12 User-friendly operation	0.2500	2.0000	1.0000	3.0000	4.0000	0.2308
G13 Light indicators	0.3333	0.2000	0.3333	1.0000	2.0000	0.0806
G14 Voice interaction	0.1429	0.1667	0.2500	0.5000	1.0000	0.0443
Inspection Criteria	${\lambda }_{max}=5.4469\hspace{1em}\hspace{1em}\hspace{1em}CI=0.1117\hspace{1em}\hspace{1em}\hspace{1em}CR=0.0998$

and

Table 14. Sub-criterion Level P4 Evaluation Indicators.

P4 Judgment Matrix	G15 Clean Interface	G16 Data Visualization	G17 User-Friendly Operation	Normalized Weight Index
G15 Clean interface	1.0000	4.0000	7.0000	0.6877
G16 Data visualization	0.2500	1.0000	4.0000	0.2344
G17 User-friendly operation	0.1429	0.2500	1.0000	0.0778
Inspection Criteria	${\lambda }_{max}=3.0775\hspace{1em}\hspace{1em}\hspace{1em}CI=0.4321\hspace{1em}\hspace{1em}\hspace{1em}CR=0.0745$

.

3.2.4. Prioritization of Demand Metrics

After the subsections in the hierarchy passed the consistency test, the comprehensive weights (w) for each indicator were determined. The relative importance of each user requirement was then calculated using Formula (7), with the final ranking of indicator importance presented in

Table 15. Ranking of Indicator Importance.

Guideline Layer P	Weight Value	Sorting	Sub-Principle Level G	Weight Value	Composite Weighting	Sorting
Functional Requirements P1	0.5818	1	Smart Ball Retrieval G1	0.4003	0.2329	1
			Smart Serve G2	0.3722	0.2165	2
			Shared Rental Model G3	0.1009	0.0587	5
			Mobility G4	0.0830	0.0434	7
			Tennis Storage G5	0.0437	0.0254	10
Esthetic needs P2	0.0471	4	Tech-inspired G6	0.4508	0.0212	11
			LED light strips G7	0.0921	0.0043	16
			Sleek design G8	0.4034	0.0190	12
			Everyday color palette G9	0.0537	0.0025	17
Interaction Requirements P3	0.1879	2	Clean interface G10	0.3687	0.0693	4
			Data visualization G11	0.2756	0.0518	6
			User-friendly operation G12	0.2308	0.0434	8
			Light indicators G13	0.0806	0.0151	13
			Voice interaction G14	0.0443	0.0083	15
Self-actualization needs P4	0.1832	3	Customized serving parameters G15	0.6877	0.1260	3
			Personalized training plans G16	0.2344	0.0429	9
			Community exchange platform G17	0.0778	0.0143	14

.

The calculation results in the table indicate that at the criteria level, functional requirements > interaction requirements > self-actualization needs > esthetic needs. At the sub-indicator level, the importance ranking of demand indicators is: Smart ball retrieval > Smart serving > Customized serving parameters > Clean interface > Shared rental model > Data visualization > Mobility > User-friendly operation > Personalized training plans > Tennis storage > Tech-inspired > Sleek design > Light indicators > Community exchange platform > Voice interaction > LED light strips > Everyday color palette.

3.3. Design Proposal for Shared Tennis Service Robots

3.3.1. Product Positioning

1. Product Positioning

The functional positioning of the shared tennis service robot is articulated across five core dimensions. First, the smart ball retrieval feature mitigates the common problem of time-consuming and labor-intensive ball collection, thereby substantially improving training efficiency through automation [35,36]. Second, the adjustable smart ball-feeding function, supported by an integrated control system and flexible feeding mechanism, allows users to customize ball speed, angle, and spin according to their skill level and training objectives, enhancing both the diversity and engagement of practice. Third, the app-based shared rental model facilitates on-demand booking and usage, lowering costs while maximizing overall product utilization [37]. Fourth, data visualization and personalized services capture and analyze training metrics to illustrate performance trends, while intelligent algorithms provide tailored training recommendations, effectively serving as a virtual “personal coach” [38,39]. Finally, multi-channel interactive feedback integrates app-based communication with visual cues, such as lighting signals, to deliver real-time operational status updates and collect user feedback, thereby enhancing the overall interactive experience [40,41].

2. Module Layout Positioning

At the hardware level, to translate identified user needs into concrete design solutions, the shared tennis service robot is composed of five primary functional modules: ball retrieval, ball serving, storage, ball transfer, and mobility. These modules operate in coordination to ensure seamless integration and efficient performance, thereby providing the structural foundation for the overall design. Based on the priority weights obtained from the Analytic Hierarchy Process (AHP) analysis, the overall layout is determined, as shown in Figure 5.

3.3.2. Design Sketch Proposal

Based on the product positioning analysis, the exterior design of the shared tennis service robot was developed. Drawing from preliminary user research, interviews, and user weight analysis results, the design prioritized functional feasibility, followed by module coordination and integration. Finally, geometric forms and detailed treatments were employed to convey design intent. Three design concepts were created, as shown in Figure 6.

As shown in Figure 6a, Scheme 1 divides the overall structure into upper and lower sections through functional zoning. The upper section comprises ball-picking, conveying, and storage mechanisms, while the lower section houses an independent ball-dispensing mechanism. The overall form features a square-within-a-circle silhouette with clean, flowing lines. As shown in Figure 6b, Design 2 adopts a more regular geometric shape, isolating the storage section for easier ball retrieval. Wavy textures are added to the glass and side partitions, enriching product details while enhancing overall texture and visual depth. As shown in Figure 6c, Option 3 also adopts a “square-meets-round” design foundation but emphasizes the independence of the storage section. Visible glass panels are added to the exterior of both the storage and delivery modules, allowing users to observe the internal operational status.

A comparative analysis of the three alternatives indicates that Option 2 (Figure 6b) provides the most effective balance between user convenience in ball retrieval and overall design coherence. In addition, the slightly squared configuration of the storage compartment enhances spatial efficiency. Accordingly, Option 2 was selected as the final design solution, with further refinements applied to its detailed features.

3.3.3. Renderings

After finalizing the design sketch proposals, this study further refined the styling and optimized details for Proposal Two to ensure it meets functional requirements while maintaining strong visual recognition. Based on preliminary research findings, most users prefer color schemes that blend technological sophistication with everyday appeal. Therefore, the shared tennis service robot’s color design should primarily feature black, white, and gray—aligning with mainstream esthetics while avoiding bright colors that could interfere with ball recognition. Simultaneously, considering the dynamic nature of tennis and the positive characteristics of users, vibrant accents like yellow and orange were strategically incorporated. This enhances the product’s approachability and vitality, creating an overall color scheme that balances technological sophistication with seamless integration into its usage environment.

Figure 7 presents the final design of the shared tennis service robot. The overall esthetic employs a predominantly black-and-white color scheme, accented with bright yellow highlights that infuse vitality into the otherwise restrained body. The design integrates geometric forms in a square-meets-round composition, with clean, dynamic lines and rounded chamfers that emphasize both functionality and visual balance. Attention to detail is reinforced through meticulous craftsmanship, including textured finishes on glass panels and raised surfaces, complemented by strategically positioned LED lighting. These elements collectively enhance the product’s technological sophistication and premium quality, while fostering an approachable and dynamic character. Furthermore, the decorative icon incorporates a stylized tennis ball silhouette as an ornamental motif, reinforcing the product’s functional identity and establishing a distinctive visual signature.

3.3.4. Functional Specifications

To ensure the feasibility and functional integrity of the design in practical applications, this study conducted structural refinement and process analysis on the key functional modules of the shared tennis service robot.

As shown in Figure 8, within the ball retrieval module, the integrated smart camera employs image recognition algorithms to locate tennis balls and track their trajectories. Combined with color segmentation and shape recognition, this achieves efficient identification [13]. Activating the motor starts the roller brush, while the ball deflector guides balls toward the brush inlet. The high-speed rotating brush then draws the balls into the internal conveying component module for subsequent collection and ball-feeding operations [36]. Nylon was selected as the material for the roller brush bristles on the CMF. It offers moderate hardness, excellent wear resistance, and strong resilience. This ensures long-term use without bristle shedding and minimizes damage to tennis balls.

To ensure efficient coordination between the ball retrieval and serving modules, this study employs a conveyor belt system with multiple baffles. This design rapidly transports tennis balls collected by the roller brush to the serving outlet, preventing balls from falling off during transit. Additionally, a transparent glass panel is integrated on the exterior of the conveyor module, facilitating real-time inspection and maintenance. This component can also be rotated open via its top-mounted gears and rotating shaft, enabling emergency intervention to resolve ball jams.

As shown in Figure 9, the primary components of the ball-serving function are two speed-adjustable differential wheels. Users can select and adjust various ball-serving parameters via the mobile app interface. The PID system controls the gears to drive the two differential wheels, achieving the user’s desired ball-serving parameters [35,36]. Additionally, the color scheme incorporates yellow accents matching the overall design, enhancing the product’s cohesiveness while preventing monotony in this module.

The tennis ball storage section features a dual-layer design to maintain the product’s unified and cohesive appearance. The outer layer is a fixed structure, molded as one piece with the main body, while the inner layer consists of a removable storage box. When the user does not require the smart ball-feeding function, the inner storage box can be removed through the opening at the top of the outer layer for convenient access and maintenance of the tennis balls.

3.3.5. Interface Design

Based on preliminary research and integrated proposals, this study established the information architecture logic of the software to further enhance the product’s interactive experience, clearly defining the functional layered structure. The core functions of the shared tennis service robot mobile application are defined as follows: robot rental, robot control and adjustment, data monitoring and recording, intelligent training reminders, and online community interaction. As shown in Figure 10, the information architecture of the APP is designed based on these functions, dividing the interface into six main modules: Guide Page, Home, Scan, Data, Community, and My Page. This structure refines the hierarchical structure of functional modules, clarifies the logical relationships between levels, streamlines interaction processes, and meets users’ expectations for an intuitive and easy-to-use interface. This logical framework also provides a solid foundation for the subsequent UI interface design.

Based on a clear information architecture, proceed with the design of the interactive interface and visual language. In the color scheme design for the Shared Tennis Service Robot app, considering users’ esthetic preference for a technological feel, blue-purple—a color frequently used in cyberpunk, a style representing futuristic esthetics—was adopted. Therefore, as shown in Figure 11, blue-purple serves as the overall color tone, and this color is also extensively applied throughout the interface design, enabling better user acceptance. To align with user expectations for a clean and intuitive interface, the overall design adopts a hybrid approach that combines flat and semi-flat styles. Interface components are streamlined, employing minimalist lines that eliminate superfluous decorative elements. This design strategy emphasizes core system functionality while enhancing the recognition of key information through the extensive use of icons.

3.3.6. Shared Rental Model Description

Against the backdrop of the sharing economy, this study introduces a shared rental model into the design of tennis service robots to reduce user costs, enhance equipment utilization, and improve the user experience. To ensure an efficient and convenient rental process, the workflow is designed around a core path of “scan—rent—use—return,” while integrating with the backend equipment management system to achieve a closed-loop service. Figure 12 illustrates the shared rental process for tennis service robots. Users initiate the service by scanning a QR code through the mobile app or by reserving specific courts and machines in advance. After completing payment and unlocking the robot, they can operate its various functions via the app and reuse features as needed. Once the session is complete, users select “End Session” in the app, prompting the robot to automatically return to its designated location and conclude the rental process.

3.4. Design Evaluation Based on Fuzzy Comprehensive Evaluation Method

Based on the results of the Analytic Hierarchy Process (AHP), a fuzzy comprehensive evaluation was conducted. An evaluation set was first defined as $V=\left\{V_1, V_2, {V_3, V}_4\right\}$, where $V_1$ represents Excellent, $V_2$ represents Good, $V_3$ represents Average, $V_4$ represents Poor. Six industry experts and product designers were invited to evaluate each indicator of the shared tennis service robot design. Scores above 90 were classified as Excellent, scores between 75 and 90 as Good, scores between 60 and 75 as Moderate, and scores between 50 and 60 as Poor. The final statistical results are presented in

Table 16. Fuzzy Evaluation Results of Shared Tennis Service Robot Design Schemes.

Evaluation Item G	Composite Weighting $\mathit{\omega }$	V1 Excellent	V2 Good	V3 Moderate	V4 Poor
Smart Ball Retrieval G1	0.2329	3	2	0	0
Smart Serve G2	0.2165	3	2	0	0
Shared Rental Model G3	0.0587	2	2	1	0
Mobility G4	0.0434	0	3	2	0
Tennis Storage G5	0.0254	0	2	3	0
Tech-inspired G6	0.0212	1	2	2	0
LED light strips G7	0.0043	0	2	2	0
Sleek design G8	0.0190	1	2	1	1
Everyday color palette G9	0.0025	2	2	1	0
Clean interface G10	0.0693	3	2	0	0
Data visualization G11	0.0518	2	0	3	0
User-friendly operation G12	0.0434	1	1	2	1
Light indicators G13	0.0151	1	1	2	1
Voice interaction G14	0.0083	2	1	1	1
Customized serving parameters G15	0.1260	5	0	0	0
Personalized training plans G16	0.0429	1	1	3	0
Community exchange platform G17	0.0143	2	0	2	1

.

Based on the data provided in Table 16, the fuzzy relationship matrix $R_1$ was derived using Equation (8).

$$R_1=\left[\begin{array}{cccc}0.6& 0.4& 0& 0\\ 0.6& 0.4& 0& 0\\ 0.4& 0.4& 0.2& 0\\ 0& 0.6& 0.4& 0\\ 0& 0.4& 0.6& 0\\ 0.2& 0.4& 0.4& 0\\ 0& 0.4& 0.4& 0\\ 0.2& 0.4& 0.2& 0.2\\ 0.4& 0.4& 0.2& 0\\ 0.6& 0.4& 0& 0\\ 0.4& 0& 0.6& 0\\ 0.2& 0.2& 0.4& 0.2\\ 0.2& 0.2& 0.4& 0.2\\ 0.4& 0.2& 0.2& 0.2\\ 1& 0& 0& 0\\ 0.2& 0.2& 0.6& 0\\ 0.4& 0& 0.4& 0.2\end{array}\right]$$

According to Equation (9), the weights $\omega$ of each indicator are multiplied by the fuzzy relationship matrix $R_1$ to derive the fuzzy comprehensive evaluation $B_1$ of the shared tennis service robot.

$$B_1=\omega \times R_1=\left(0.5198, 0.3089, 0.1464, 0.0200\right)$$

The membership degrees for the rating categories of Excellent, Good, Average, and Poor in this scheme are 51.98%, 30.89%, 14.64%, and 2%, respectively. According to the maximum membership principle, the shared tennis service robot designed in this study is therefore evaluated as Excellent. By assigning scores of 95, 85, 75, and 65 to V₁, V₂, V₃, V₄ respectively, and letting $Y={\left(95, 85, 75, 65\right)}^T$, the following result is obtained from Equation (10):

$$Z=B\circ Y=\left(95\times 0.5198+85\times 0.3089+75\times 0.1464+65\times 0.0200\right)=87.8349$$

In summary, the shared tennis service robot developed in this study was validated through objective evaluation metrics that combined qualitative and quantitative methods while comprehensively accounting for user needs. The design achieved a final score of 87.8349, placing it within the “Good” to “Excellent” range. These findings indicate that the proposed shared tennis service robot demonstrates strong effectiveness in meeting user requirements.

4. Discussion

Current research on tennis robots has largely concentrated on single functions, such as automatic ball feeding or ball retrieval, with relatively limited efforts to integrate both capabilities. Existing literature primarily emphasizes the realization of mechanical functions and improvements in localized performance, such as ball trajectory accuracy, recognition precision, or efficiency gains in individual modules. However, research on overall user experience, system coordination, and shared application models remains insufficient. For instance, Ye et al. proposed a tennis serve training system based on a local attention convolutional neural network, focusing on recognition accuracy and performance optimization during the serve process. Yet, their research scope was primarily confined to a single serve function [12]. Huang et al. developed a vision-based intelligent tennis ball-retrieval robot that enhanced automation in ball collection, yet did not address shared usage scenarios or multifunctional integrated design [5]. This study expands the research perspective on multifunctional collaboration and service applications for smart sports equipment by integrating ball retrieval, ball serving, and shared rental mechanisms within a single product. It provides new technical pathways and application models for designing sports service robots, effectively addressing the dual challenges of enhancing training efficiency and reducing usage costs. In contrast to prior research that primarily emphasized mechanical reliability or single-function improvements, this work simultaneously prioritizes user experience, interaction design, and practical operability. It broadens the research perspective of intelligent sports equipment and demonstrates clear innovation in integrating system functionality with the shared economy model.

Compared with previous studies, this study employs the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE) to establish a systematic and scientific approach for the design of shared tennis service robots. Using AHP, user requirements were quantified and prioritized across four dimensions—functional needs, esthetic preferences, interactive expectations, and self-directed desires—resulting in a hierarchical model comprising 17 indicators. This framework highlighted the pivotal roles of “intelligent ball retrieval”, “intelligent ball serving”, and “customizable serving parameters” in guiding the product’s direction. In implementing this methodology, the study invited eight cross-disciplinary experts to participate in the evaluation. Geometric averaging and consistency checks (CR ≤ 0.1) were employed to mitigate the impact of individual differences on the results, thereby enhancing the stability and objectivity of the conclusions. Given the relatively limited size of the expert panel, while this approach provides clear priority guidance, the robustness of the weightings can be further validated and refined through subsequent expansion of the expert group or integration with additional empirical user behavior data. The subsequent application of FCE enabled a comprehensive evaluation of these prioritized requirements, thereby validating both the rationality and feasibility of the proposed design. By integrating the AHP–FCE methodology, this study not only quantified user requirements and established their priority rankings but also mapped relationships from user needs and design metrics to product structural modules. This approach provides a replicable pathway for the systematic design of shared sports equipment products. Compared to single-function or subjective design methods, this methodology facilitates the formation of clear technical roadmaps and functional implementation strategies during the early design phase, thereby shortening R&D cycles and enhancing the relevance and practicality of designs. From a broader perspective, this study integrates user-centered design methodologies with systematic evaluation techniques, contributing to the advancement of sports service robotics. The findings provide valuable insights for the design and assessment of future shared sports equipment, emphasizing the potential of the sharing economy model to reduce barriers to equipment access and improve resource utilization efficiency.

In terms of engineering implementation, this study proposes a modular design solution featuring roller brush collection, conveyor belt transport, differential wheel ball delivery, and dual-layer storage, providing a clear path for subsequent technical realization. For the core “intelligent ball retrieval” function, target detection is achieved via RGB cameras and lightweight recognition algorithms. Autonomous navigation is enabled through IMU and odometer positioning, with an interface reserved for SLAM integration. The ball-dispensing module employs differential wheels and PID control for adjustable ball-release parameters. The storage module adopts a dual-layer structure to enhance maintenance and operational convenience. Key engineering challenges include outdoor environmental impacts on recognition accuracy, high field randomness, device endurance and energy consumption control difficulties, and mechanical reliability under high-frequency use. To address these, technologies such as lightweight visual algorithms, ORB-SLAM2 positioning, autonomous navigation, and modular motor control [13,35,36] can be leveraged to enhance system stability and scalability. Integrating these technical approaches with the design framework proposed in this study can effectively shorten the cycle from conceptual design to engineering implementation while improving the robot’s adaptability and reliability in real-world scenarios.

Due to limitations in the scope and timeframe of this study, the sample primarily focused on the Chinese region, which to some extent restricts the external generalizability of the findings. Preferences for the “smart ball feeding” feature, shared usage models, and price sensitivity may vary across different cultural, economic, and sports accessibility contexts, directly impacting the prioritization of design elements in Table 15. For instance, regions with higher tennis participation may exhibit stronger acceptance of shared models, while areas with relatively limited sports resources might assign greater weight to equipment cost and accessibility. Consequently, the feature hierarchy constructed in this study is best suited for the market environment represented by the current sample characteristics. Localized adjustments and refinements may be necessary when applying these findings to other regions.

Future research may validate the stability and applicability of the model by expanding the sample size of experts and users and incorporating cross-regional and cross-cultural data. In addition, as this study remains at the prototype and evaluation stage, it has not yet undergone large-scale validation in real-world scenarios. The stability and reliability of the system under complex court conditions and frequent use therefore require further empirical testing.

5. Conclusions

This research utilizes the AHP–FCE method to construct a design assessment model for shared tennis service robots. The Analytic Hierarchy Process (AHP) was initially used to systematically categorize user requirements, yielding 17 evaluation indicators across four dimensions: functionality, interaction, esthetics, and self-service (e.g., intelligent ball retrieval with a weight of 0.2329, intelligent ball serving with a weight of 0.2165). Comprehensive weights and rankings for each indicator were then calculated. Subsequently, the Fuzzy Comprehensive Evaluation (FCE) method was used to determine the membership degrees of each indicator for the proposed design. By combining these results with a weighted average model, a comprehensive score of 87.83 was obtained. This demonstrates that the AHP–FCE approach effectively integrates qualitative analysis with quantitative validation, translating user requirements into concrete design elements.

This study not only proposes a shared service robot design tailored for tennis scenarios but also integrates data visualization with personalized training functions, offering a practical pathway to enhance user motivation and engagement in sports activities. It simultaneously provides a reusable design evaluation framework for the design and assessment of similar service products in badminton, table tennis, basketball, and other sports, thereby expanding the application scenarios of sports service robots within the context of the sharing economy.

While this study is limited by sample size and user diversity, its methodology has been validated in terms of both reusability and applicability. Future research could optimize design strategies by expanding sample sizes, conducting cross-regional and cross-cultural user research, and incorporating artificial intelligence with adaptive control algorithms, thereby further enhancing user experience and contributing valuable insights to the innovation of sports service robots within the sharing economy.