The rise of the low-altitude economy represents a transformative shift in global aviation and economic development, positioning itself as a critical component of strategic emerging industries and a manifestation of new quality productive forces. At the heart of this transformation lies the unmanned aerial vehicle (UAV), or drone, industry. As a core driver of the low-altitude economy, the China drone sector exhibits tremendous potential and broad application prospects across logistics, agriculture, urban governance, and emergency response, significantly enhancing airspace utilization efficiency and socio-economic productivity. Understanding the spatial distribution and evolutionary dynamics of this industry is crucial for optimizing regional industrial layouts and fostering the sustainable development of the low-altitude economy. This article, based on an in-depth analysis of enterprise data, aims to systematically reveal the spatiotemporal evolution characteristics and underlying driving mechanisms of the China drone industry.
The evolution of industrial spatial patterns and their driving mechanisms has long been a central theme in economic geography. Classical location theories, such as Weber’s industrial location theory, emphasized static factors like transportation costs and resource endowment. However, the rise of knowledge-intensive industries like the China drone sector necessitates frameworks that incorporate innovation resources, institutional environments, and network linkages. Agglomeration theory explains the advantages of clustering through knowledge spillovers and shared labor markets, while evolutionary economic geography highlights the path-dependent nature of industrial development, where regional pre-existing capabilities shape the trajectory of emerging sectors. This article integrates these perspectives to construct a comprehensive analytical framework for the China drone industry.
Our analysis is grounded in data encompassing 45,070 drone-related enterprises across 280 prefecture-level cities in China from 2010 to 2022. The enterprises are classified into three segments of the industry chain: Upstream (R&D, Design, Testing), Midstream (Manufacturing, Integration), and Downstream (Operations, Services). The classification is performed using a combination of keyword screening and manual verification based on business scope descriptions, as summarized in the table below.
| Industry Chain Segment | Description & Keywords (Examples) | Enterprise Count |
|---|---|---|
| Upstream | System design, R&D, algorithm development, flight testing, chip/sensor development. | 21,112 |
| Midstream | Manufacturing, assembly, hardware integration, production of components (motors, frames, flight controllers). | 24,663 |
| Downstream | Operation services, aerial surveying, agricultural plant protection, logistics, training, data processing. | 31,473 |
Spatial analysis methods form the core of our investigation. We employ Kernel Density Estimation (KDE) to visualize the intensity and evolution of the China drone industry’s spatial agglomeration. The KDE function is given by:
$$f(y) = \frac{1}{nh}\sum_{z=1}^{n} k\left(\frac{y – y_z}{h}\right)$$
where \(k(\cdot)\) is the kernel function, \(|y – y_z|\) is the distance from estimation point \(y\) to sample point \(y_z\), \(h\) is the bandwidth, and \(n\) is the number of enterprises. Global Moran’s I index is used to measure the overall spatial autocorrelation:
$$\text{Moran’s I} = \frac{N \sum_{i=1}^{N} \sum_{j=1}^{N} W_{ij}(x_i – \bar{x})(x_j – \bar{x})}{(\sum_{i=1}^{N} \sum_{j=1}^{N} W_{ij}) \sum_{i=1}^{N} (x_i – \bar{x})^2}$$
where \(x_i\) and \(x_j\) are the number of enterprises in cities \(i\) and \(j\), \(\bar{x}\) is the mean, and \(W_{ij}\) is the spatial weight matrix. Finally, the Geodetector model is utilized to identify driving factors and their interactions. The factor detection \(q\)-statistic measures the explanatory power of a factor \(X\) on the spatial heterogeneity of enterprise distribution \(Y\):
$$q = 1 – \frac{1}{N\sigma^2} \sum_{p=1}^{L} N_p \sigma_p^2 = 1 – \frac{SSW}{SST}$$
where \(N\) and \(N_p\) are the number of units in the entire region and sub-region \(p\), \(L\) is the number of strata, \(\sigma^2\) and \(\sigma_p^2\) are the variances of \(Y\) in the entire region and sub-region \(p\), \(SSW\) is within-sum of squares, and \(SST\) is total sum of squares. Interaction detection assesses whether two factors \(X_a\) and \(X_b\) enhance or weaken each other’s influence.
The spatial pattern of the China drone industry exhibits pronounced agglomeration characteristics. Global Moran’s I indices from 2010 to 2022 are consistently positive and statistically significant, confirming a non-random, clustered distribution. The value experienced a slight decline post-2016, likely due to national policy encouragement leading to broader diffusion, before rising again to a higher level in 2022, indicating a reinforcement of spatial clustering.
| Year | Moran’s I | Z-score | P-value |
|---|---|---|---|
| 2010 | 0.112 | 3.820 | 0.000 |
| 2016 | 0.093 | 3.924 | 0.000 |
| 2022 | 0.144 | 4.724 | 0.000 |
The core agglomerations are firmly established in China’s major city clusters: the Pearl River Delta (PRD), the Yangtze River Delta (YRD), and the Beijing-Tianjin-Hebei (BTH) region. The evolutionary trajectory shows a clear shift from initial “point distribution” in these core areas around 2010 towards a “multi-core agglomeration” pattern by 2022. While the PRD, led by Shenzhen, maintains the highest density core, secondary high-density nuclei have solidified in the YRD and multiplied within the BTH region. This evolution underscores the deepening and spatial diversification of the China drone industrial base. Furthermore, significant spatial differentiation exists along the industry chain. Upstream and midstream segments show even stronger concentration in the core coastal megaregions and provincial capitals, reliant on advanced innovation and manufacturing ecosystems. The downstream segment, being more application-oriented, demonstrates a relatively wider distribution, penetrating into inland cities with specific demand scenarios like agriculture and security, though often lacking the complete industrial chain.
To decipher the drivers behind this spatial pattern, we construct an indicator system based on integrated theoretical frameworks. The drivers are categorized into six dimensions: Economic Foundation, Application Scenario, Institutional Support, Industrial Agglomeration, Industrial Linkage, and Innovation Drive. Each dimension is measured by specific proxy variables, as detailed in the analysis.
The Geodetector results for factor influence (q-statistic) reveal a hierarchy of drivers shaping the China drone industry landscape. The most powerful single factor is Innovation Output (e.g., patent applications), with a q-value of 0.404, underscoring the fundamental role of technological capability. This is followed closely by factors representing Industrial Linkage, specifically the density of Software & IT services (q=0.365) and Scientific & Technical services (q=0.324). These findings highlight that the growth of the China drone sector is not isolated but deeply embedded in a broader ecosystem of knowledge-intensive services and advanced manufacturing. Application Scenario factors, particularly Economic Density (q=0.288), also show strong explanatory power, indicating that market potential and the concentration of economic activity are critical for attracting and sustaining drone enterprises. The results validate that while traditional factors like Economic Foundation and Institutional Support remain relevant, the spatial dynamics of this emerging industry are predominantly governed by innovation, related industrial variety, and market accessibility.
| Dimension | Representative Factor | q-statistic (Overall) | Interpretation |
|---|---|---|---|
| Innovation Drive | Innovation Output (Patents) | 0.404 | Primary driver; core technological capability. |
| Industrial Linkage | Software & IT Services | 0.365 | Key enabling ecosystem; provides essential technical support. |
| Application Scenario | Economic Density | 0.288 | Critical for market formation and service demand. |
| Economic Foundation | Per Capita GDP | 0.127 | Provides essential underlying economic capacity. |
| Institutional Support | Government Fiscal Capacity | 0.115 | Facilitates through policy and financial support. |
The interaction detection analysis yields even more insightful results. It demonstrates that the combined effect of any two factors is greater than their individual influence, exhibiting nonlinear enhancement or bi-enhancement. The synergy between Innovation Output and other factors is particularly potent. For instance, the interaction between Innovation Output and Human Capital yields a q-value of 0.596, and its interaction with Electronic Manufacturing reaches 0.676. This implies that innovation’s impact on the China drone cluster is massively amplified by the presence of a skilled talent pool and a sophisticated hardware manufacturing base. Similarly, Economic Density interacts strongly with Human Capital (q=0.737), suggesting that economically dense areas are successful because they co-agglomerate both market demand and the human resources needed to serve it. These interactions can be conceptually modeled. The location attractiveness \( A_i \) for drone enterprises in city \( i \) can be expressed as a function of these interacting drivers:
$$A_i = \alpha \cdot \underbrace{(INNO_i \cdot HC_i)}_{\text{Innovation-Talent Synergy}} + \beta \cdot \underbrace{(SCEN_i \cdot LINK_i)}_{\text{Market-Ecosystem Synergy}} + \gamma \cdot INST_i + \epsilon_i$$
where \(INNO_i\) is innovation capacity, \(HC_i\) is human capital, \(SCEN_i\) is application scenario strength, \(LINK_i\) is industrial linkage, \(INST_i\) is institutional support, and \(\alpha, \beta, \gamma\) are coefficients.
The differential influence of drivers across the industry chain segments is notable. Upstream segments are most sensitive to Innovation Drive and Industrial Linkage to scientific services. Midstream manufacturing relies heavily on the synergy between Industrial Linkage (both hardware and software) and Agglomeration economies. Downstream services are primarily driven by Application Scenarios and, to a significant extent, by Innovation Drive, which provides the technological solutions applied in the field. This segmentation suggests that regional strategies for developing the China drone industry must be tailored: regions aiming for high-value R&D must prioritize innovation ecosystems, while those focusing on applications can leverage local market needs and adapt existing technologies.
The findings place the evolution of the China drone industry within broader theoretical discourses. The strong role of Innovation Drive and Industrial Linkage resonates with the knowledge-based and evolutionary turns in economic geography, emphasizing the importance of regional knowledge bases, related variety, and path dependency. The significant influence of Application Scenarios extends traditional market-seeking location logic by incorporating the critical dimension of technology adoption potential and solution-space fit, which is paramount for emerging industries. The persistent, though secondary, role of Economic Foundation and Institutional Support confirms that even highly innovative sectors do not operate in a vacuum; they require a solid macroeconomic foundation and supportive governance frameworks to flourish. The China drone industry’s spatial pattern is thus a result of a complex, multi-scalar process where localized innovation synergies, embedded within regional industrial ecosystems, are facilitated by broader economic and institutional conditions.
This analysis, while comprehensive, has limitations. The use of enterprise count, rather than metrics like revenue or employment, may not fully capture the economic weight and productivity of clusters. Future research could incorporate such data, along with network analysis of inter-firm collaborations and innovation linkages, to provide a more granular understanding of the China drone industry’s internal structure and knowledge flows. Tracking the evolution of firm entry, exit, and growth within these spatial patterns would also offer deeper insights into the dynamic competitive landscape.
In conclusion, the spatial evolution of the China drone industry is characterized by strengthening agglomeration within leading megaregions and a gradual diffusion shaped by a clear hierarchy of drivers. Innovation capability, synergies with related knowledge-intensive industries, and localized application demand are the principal forces configuring the industrial map. This has profound implications for policy and regional development strategy. For core regions like the PRD, YRD, and BTH, the focus should be on maintaining innovation leadership by fostering cutting-edge R&D, strengthening cross-industry knowledge spillovers, and developing advanced testing and certification infrastructures. For emerging regions, strategy should pivot towards identifying and developing niche application scenarios (e.g., agricultural drones in major farming provinces, inspection drones in industrial corridors) and building connections to innovation hubs for technology transfer. Nationally, policy must continue to streamline low-altitude airspace management, develop standardized regulations, and promote interoperability to create a cohesive environment for the China drone industry to innovate and scale. By adopting such a spatially and sectorally differentiated approach, China can optimize the layout of this strategic industry, foster balanced regional development within the low-altitude economy, and solidify its global competitive edge in the drone sector.