Introduction: Three Years of Hustle and Bustle in the Sky, but the Real Threshold Has Not Yet Been Reached
Over the past three years, the most intuitive change in China's low-altitude economy is the sudden influx of "flying machines" into the public eye. At the Zhuhai Airshow, tilt-rotor aircraft completed transition flights in public outside the exhibition hall; in livestream rooms in urban commercial districts, an "air taxi" was bought in full; at the CIIE and the Low-Altitude Economy Expo, six-seat compound-wing real aircraft were surrounded by crowds; in official local government promotions, "maiden flight," "roll-out," and "global debut" have become high-frequency words. The enthusiasm of capital, media, and local investment promotion has been ignited simultaneously.
I. Three Intriguing Contrasts
But if we shift the lens from press conferences back to regulatory documents, we will see three somewhat discordant contrasts.
The first contrast: The clamor of exhibitions versus the silence of airworthiness announcements. There are 15 publicly unveiled domestic eVTOL projects, but only 2 complete certification records can be found. On one side are intensive "maiden flights," "transition flights," and "global debuts"; on the other side are the very few Type Certificates in the CAAC airworthiness announcement column. Between "being able to fly" and "being able to legally carry passengers" lies an entire set of compliance verification, not just a single successful test flight.
The second contrast: The hustle and bustle of order numbers versus the scarcity of delivery records. Volante disclosed 260 orders, ZG-ONE has thousand-level intentions, AR-E800 secured 160 confirmed and intended orders, and the Land Aircraft Carrier has pre-sale and overseas order disclosures—adding up to a scale of thousands. However, the number of cases where single aircraft delivery has been truly completed and can be counted into the fleet is屈指可数 (few and far between). Intended orders, framework agreements, confirmed orders, and delivery—these four terms are often compressed into one word in communication: "orders."
The third contrast: The blooming of configurations versus the high convergence of parameters. Multi-rotor, compound-wing, tilt-rotor, semi-tilt, and detachable flying cars—the routes look diverse. But once the parameters are laid out, passenger models are almost all squeezed into the same window: maximum takeoff weight of 2-2.6 tons, range of 200-300 km, and cruise speed of 200-250 km/h. This is not because enterprises are copying each other's homework, but because it is the only solution interval squeezed out by battery energy density, noise constraints, takeoff and landing site sizes, and certification difficulty. The "blooming" of configurations is, to some extent, different compromise postures under the same physical ceiling.
II. An Overall Judgment: The Three-Stage Theory
Looking at these three contrasts together, we can make a more accurate judgment on the current industrial stage of domestic eVTOL:
The first stage, "building it," is basically over,
The second stage, "certifying it," is under tough attack,
The third stage, "using it and earning it back," has not yet truly begun.
Three governance philosophies: Beijing consolidates control through legislation, Shanghai and Guangzhou consolidate release through legislation, and Hefei and Chengdu are exchanging space in the pursuit of pilot programs.
This sentence has three specific meanings.
First, "building a flying eVTOL" is no longer a barrier. The existence of 15 projects is evidence in itself. What is truly scarce is the ability to freeze the design into a certifiable configuration and continuously produce consistent products.
Second, the current bottleneck is not technology, but compliance verification and quality systems. Most projects are stuck in the "airworthiness acceptance—compliance verification" interval. What is lacking is not creativity and engineering capability, but a systematic, traceable, and regulator-acceptable chain of evidence.
Third, the validation of business models has not yet truly begun. Of the two certified models, one mainly operates low-altitude cultural tourism and designated routes, and the other operates cargo and offshore operations. Neither has yet formed replicable and auditable large-scale operational data. This means the entire industry is still in the "investment period" rather than the "return period."
III. A Ruler Running Through the Entire Low-Altitude Cycle
From an eVTOL's debut to its truly legal, stable, and replicable passenger carrying, there are three doors in between: Type Certificate (TC), Production Certificate (PC), and Airworthiness Certificate (AC). TC proves the design is safe, PC proves the production line can continuously build equally safe aircraft, and AC proves the specific aircraft in front of you is in an airworthy condition. Only when all three are complete is the qualification foundation for mass production and operation established.
Measured by this ruler, as of September 2026, among the 15 publicly unveiled projects, only 2 have completed the three-certificate closed loop: EHang's EH216-S (unmanned passenger-carrying) and AutoFlight's V2000CG Kerry Gull (ton-class unmanned cargo). The remaining 13 are distributed across four levels: "airworthiness acceptance," "compliance verification," "full-scale test flight," and "just completed debut."
This gap is not bad news, nor is it industrial stagnation. Its more accurate meaning is: Domestic eVTOLs have collectively crossed the stage of "whether they can be built and whether they can fly," and are now collectively tackling the three things: "whether they can be proven safe, whether they can be continuously built, and whether they can make money." And these three things precisely determine who can survive the next round of differentiation.
Chapter 1 Current Status: 15 Names, 2 Completed Paths
Let's first put all objects on the table. The following table is sorted by "progress maturity" rather than enterprise size—this is the most useful ruler for understanding this industry.
First, configurations are highly concentrated in "compound-wing + tilt-rotor." Among the 15 eVTOL projects, 6 are compound-wing, 3 are tilt-rotor, 3 are multi-rotor, and the rest belong to semi-tilt compound-wing, detachable flying cars, and new products with incomplete configuration disclosures. This shows that the industry has generally accepted the same design constraint: vertical takeoff and landing is responsible for "being able to take off from any flat ground," while fixed-wing or tilt propulsion is responsible for "flying far and flying fast." Pure multi-rotors can fly, but the physical ceiling for range and speed is too low to support intercity missions.
Second, maturity shows a typical "olive-shaped" distribution. The top 2 are certified, the bottom 2 have just debuted, and the middle 11 are densely squeezed in the "engineering verification—airworthiness certification" stage. This shape shows that the industry has no gap, and the echelons are continuous; but it also shows that the vast majority of enterprises have not yet obtained the final ticket, and the first echelon has not yet truly pulled away.
Third, parameters are highly converged. Passenger models mostly fall into the window of "2-2.6 tons maximum takeoff weight, 200-300 km range, and 200-250 km/h cruise speed." This is the technical balance point of the industry at this stage, meaning that differentiation can no longer be formed solely by parameters—the real differences will appear in certification progress, mission scenarios, and unit costs.
Figure 1: Compound-wing is currently the most concentrated technical route for domestic eVTOL
Figure 2: Among passenger and passenger-cargo universal models included in maturity comparison
Summary: 15 eVTOL projects, 4 maturity levels, 2 completed paths. This pattern shows that the industry has completed "from 0 to 1" and is experiencing the long validation period before "from 1 to 100." What is truly worth asking next is not "who else is going to release new products," but "among these 11 projects stuck in the middle, who can finish the last leg first."
Chapter 2 Technology: Four Routes, Each with Its Own Physical Ceiling
Configuration is not an aesthetic choice, but the result of mission boundaries, physical laws, and certification difficulty jointly determining. Only by understanding this can we understand why compound-wing is the most crowded, tilt-rotor is the slowest, and multi-rotor was certified the earliest.
I. Why Configurations Diverge: Three Constraints
The design of any eVTOL is essentially dealing with three conflicting constraints simultaneously:
Vertical takeoff and landing requirements: There must be a system capable of generating sufficient lift at zero speed, which is often dead weight during cruise.
Cruise efficiency requirements: There must be fixed-wing and efficient propulsion, otherwise range and speed cannot go up.
Safety redundancy requirements: No single-point failure can lead to catastrophic consequences, which means power, flight control, and energy systems must have redundant backups, and redundancy means weight.
Battery energy density determines the tightness of these three constraints: when energy density is insufficient, the weight cost paid for redundancy and vertical takeoff and landing will directly eat up the range. This also explains why all passenger models eventually converge to similar parameter windows—it is not a coincidence, but the same optimal solution under the same set of constraints.
The four technical routes are essentially four different compromise methods for these three constraints.
II. Compound-Wing: The Most Crowded Track, and Also the Breakthrough for Cargo
The basic idea of compound-wing (lift+cruise) is very simple: Use an independent lift system during vertical takeoff and landing, and rely on the fixed wing for lift and the propellers for thrust after transitioning to cruise. The two systems perform their respective duties without accommodating each other. The cost is more components, more complex transition control, and a larger workload for airworthiness compliance verification. But precisely because this "clumsy method" has clear engineering boundaries, it has become the most crowded track in domestic eVTOL.
V2000CG Kerry Gull: A Complete Closed Loop for Ton-Class Cargo
AutoFlight's V2000CG is the model with the most complete airworthiness evidence in the compound-wing route. The Civil Aviation Administration of China disclosed that the East China Regional Administration issued the Type Certificate for the V2000CG unmanned aircraft system to Shanghai AutoFlight Aviation Technology in March 2024; subsequently, the model further obtained the Production Certificate and Airworthiness Certificate, and achieved single aircraft delivery to Heli Chuangxing Intelligence in July 2025. The parameters are: maximum takeoff weight of 2000 kg, payload of about 400 kg, range of about 200 km, and speed of about 200 km/h.
The significance of this model lies in opening up the certification path for "ton-class and above unmanned cargo aircraft." Compared to passenger carrying, cargo does not need to consider passenger protection, cabin environment, escape, and in-flight services, making the certification object more focused; compared to small drones, it truly possesses commercial payload capacity. This makes V2000CG the first "legally usable" domestic ton-class eVTOL in scenarios such as logistics, offshore platform operation and maintenance, and emergency material transportation.
But "complete three certificates" does not mean "already profitable." Whether sustainable cash flow can be formed depends on route density, payload utilization rate, maintenance system, airspace coordination capability, and the real cost per sortie. Airworthiness is the entry ticket, not the income statement.
Shengshilong: Shared Platform, but Independent Certification Required
AutoFlight's passenger model V2000EM "Shengshilong" shares the compound-wing platform with V2000CG, carrying up to 6 people, with a maximum takeoff weight of about 2200 kg and a range of about 200 km. It is positioned for urban commuting, intercity travel, and low-altitude cultural tourism. It has currently entered the compliance verification stage of airworthiness certification, and the enterprise expects to complete certification in 2027.
There is a logic that must be clarified here: cargo model certification cannot replace passenger model certification. A shared platform can reduce the costs of technical experience, supply chain, and test flight data, but the requirements for passenger configuration in terms of passenger safety, system redundancy, cabin environment, and operation rules are an entirely different set. Moreover, "expected to be certified in 2027" belongs to enterprise planning, not a regulatory commitment, and should not be cited as an established fact.
LE200, VE25-100, and M1: Three Progress Lines of Mainstream Compound-Wing
These three represent the mainstream maturity of domestic compound-wing projects. All have crossed the "prototype display" stage, but the bottlenecks are different.
Lanyi Aviation's LE200 has the most regular R&D path: In 2023, it completed the scale-down verification aircraft and test flights; in 2024, it completed the full-scale engineering verification aircraft; in 2025, it continued to conduct full-scale test flights. The TC application was accepted on April 26, 2025, and the first meeting of the type certification review was held in December 2025. The parameters are: 1 pilot + maximum 5 passengers, maximum takeoff weight of about 2600 kg, cruise speed of 200 km/h, basic range of 200 km, and maximum range of 320 km. The convening of the first review meeting means the project has officially entered concentrated review, but it does not mean certification is imminent.
Volante's VE25-100 is the most active at the order level: Takeoff weight is about 2.5 tons, design range is 200-400 km, cruise speed is 235 km/h. In September 2023, it became one of the first passenger-carrying eVTOL projects in the East China region to receive airworthiness acceptance. At the 2026 International Low-Altitude Economy Expo, the enterprise disclosed obtaining 260 orders, of which 60 are from Yuntian Group Indonesia and 200 are from Minsheng Financial Leasing. These orders are worth paying attention to, but must be understood as "disclosed orders/intended orders"—public materials do not have a unified delivery definition, deposit terms, and timetable, and cannot be counted into the delivered fleet.
UFUTURE's M1 has solid engineering progress: The Yancheng Economic Development Zone disclosed that in July 2026, M1 completed the transition test from hover, vertical takeoff to fixed-wing high-speed cruise; maximum takeoff weight is 2.5 tons, maximum payload is 700 kg, pure electric range is 250 km, cruise speed is 200 km/h, adopting 16 sets of vertical rotors and 4 sets of front-pull propulsion propellers. During the same period, its 2-ton-class aerial medical ambulance mission verification aircraft jointly developed with Shanghai Ruijin Hospital made its debut.
Medical rescue is one of the most easily accepted scenarios for eVTOL in the early stage: high mission value, relatively insensitive to cost, and good social acceptance. But the debut of a mission verification aircraft does not equal model certification. M1's passenger certification target is still placed around 2028, and at this stage, it should be classified as "engineering and mission verification in progress."
AT8000: Pushing the Competition Scale to Ton-Class Heavy Load
Muyu Tian Aviation's AT8000 is the "alien" among these models. The Wuxi High-tech Zone disclosed that the aircraft's maximum takeoff weight reaches 8 tons, adopting a semi-tilt configuration of "18 lift rotors, of which 4 are tiltable, plus 2 propulsion propellers"; in range-extender mode, the effective payload is 3500 kg, full-load range is 1000 km, and cruise speed is 252 km/h.
In July 2026, AT8000 completed the roll-out of the complete fuselage. The enterprise has completed the scale-down aircraft verification flight, and subsequent steps still need to promote final assembly, system integration, ground testing, and maiden flight. Therefore, its current accurate positioning is "fuselage roll-out + scale-down verification completed," and it cannot be written as "maiden flight completed" or "airworthiness acceptance obtained."
Its real challenge is not just the payload numbers. An 8-ton-class complete aircraft requires the establishment of a completely new certification method, and its economy depends on whether tasks such as high-timeliness logistics, island and mountain area supply, and engineering hoisting have sufficient density. Maximum takeoff weight is a parameter, not competitiveness.
Figure 3: Except for heavy-load models, domestic passenger-carrying eVTOLs are highly concentrated in the parameter window of "2-2.6 tons, 200-300 km"
Figure 4: Range spans two orders of magnitude: from the 20 km cultural tourism circle to the 1000 km heavy-load range extender
III. Tilt-Rotor: Highest Performance Ceiling, Highest Verification Cost
Tilt-rotor is the configuration in eVTOL that "looks the most elegant but is the most difficult to build." Its idea is to let the same set of propellers (or rotors) face upwards during vertical takeoff and landing, and turn forward during cruise, thereby combining the takeoff and landing flexibility of multi-rotors with the cruise efficiency of fixed wings. Theoretically, it is most suitable for urban agglomeration and intercity missions.
The engineering cost is: the mechanical reliability of the tilt mechanism itself, the flight control laws during the transition phase, the aerodynamic coupling between rotors and wings, and the redundant safety after single-point failure must all be repeatedly proven within the entire flight envelope. In other words, tilt-rotor turns the question of "whether it can fly" into the question of "whether it can be proven safe in all states." This also explains why the three tilt-rotor models included in this report—AE200, E20, and ZG-T6—have all not entered the "certified" ranks.
AE200: Currently the Model with the Most Forward Engineering Progress
AEROFUGA's AE200 relies on the Geely Technology ecosystem, adopts a tilt-rotor pure electric configuration, and the cabin is a manned passenger-carrying vertical takeoff and landing aircraft with a flexible layout of 3-6 seats, with a range of 200 km. It has two key nodes: In June 2024, the team successfully completed key technical research, and AEROFUGA became the "world's second and domestic first" OEM to complete the "full-size, full-weight, full-envelope" tilt transition test flight of a 2.5-ton-class eVTOL. In September 2025, the first mass-produced aircraft of the first manned passenger-carrying eVTOL, AE200-100, was successfully rolled out. The Southwest Regional Administration of CAAC had previously accepted the AE200-100 Type Certificate application and established a project review team, issuing China's first Type Certification Acceptance Notice for a manned passenger-carrying eVTOL.
The nine words "full-size, full-weight, full-envelope" carry a lot of weight. Many projects have only completed scale-down verification or unloaded transition, while AE200 has proven the flyability of the configuration under real weight and complete envelope. But "first acceptance notice" and "taking the lead in completing transition flight" are both progress, not qualifications. AE200 still needs to complete the full compliance verification to obtain the TC.
Here, an important cognitive correction is also needed: The airworthiness difficulty of manned and unmanned products cannot be directly compared. EH216-S has been certified, while AE200 has not, but this does not mean the latter is "backward." The former is an unmanned passenger-carrying aircraft, and the latter is a manned passenger-carrying passenger aircraft. The two belong to different mission systems in terms of flight rules, pilot responsibilities, man-machine interfaces, and system redundancy. The certification basis is different, and the timelines are naturally incomparable.
E20 and ZG-T6: Both at the Entrance of Certification, with Different Depths
TIME's E20 is a 5-seat tilt-rotor. The TC application was accepted by the East China Regional Administration of CAAC in October 2023, and has entered the review coordination and compliance verification stages such as G-1, G-2, and compliance plans. Public materials have different calibers for its cruise speed and maximum range. This article does not list precise values that have not been verified under unified test conditions—when there are multiple claims for the same model, the correct approach is to list them side by side, rather than picking a "seemingly reasonable" number.
Zero Gravity's ZG-T6 adopts a full-tilt six-rotor configuration, with a maximum takeoff weight of about 2500 kg, carrying 6 people, a cruise speed of 250 km/h, and a range of 300 km. The TC application was accepted in February 2026, and the full-scale engineering prototype has been assembled and rolled out, conducting test flights. Therefore, it should be classified in the "full-scale prototype testing and airworthiness acceptance stage."
LE200, VE25-100, and M1: Three Progress Lines of Mainstream Compound-Wing
These three represent the mainstream maturity of domestic compound-wing projects. All have crossed the "prototype display" stage, but the bottlenecks are different.
The common point of the two is that they have both obtained the admission ticket for certification. The difference lies in the depth of enterprise disclosure and the current flight stage. What determines whether they can stay in the first echelon is not the timing of the press conference, but these three things: the speed of flight envelope expansion, whether the certification basis can be frozen, and whether the compliance evidence can be completed on time.
IV. Detachable Flying Car: Most Recognizable, and Also Most in Need of Cross-Boundary Closed Loop
XPENG AEROHT's "Land Aircraft Carrier" is the domestic flying car project with the highest public awareness, and also the most special in configuration: the land vehicle and the flying vehicle (X3-F) can be separated and combined, and the flying vehicle is a six-rotor dual-duct electric configuration.
The public milestones of the "Land Aircraft Carrier" are: first public debut in October 2023, public manned flight at the 2024 Zhuhai Airshow, in March 2025, the Civil Aviation Administration of China released the draft for comments on special conditions for X3-F, and the Production Certificate application was accepted. As of September 2026, existing evidence is insufficient to determine that it has obtained TC, PC, or AC.
The release of the draft for comments on special conditions is actually a positive signal. It shows that the review agency has begun to promote rule docking for the specific design features of this flying vehicle—for a completely new configuration without existing airworthiness standards, this step is a necessary prerequisite. The acceptance of the PC application indicates that the mass production system has begun to enter the scope of review. But neither is final certification.
The real risks of detachable flying cars are concentrated on the interfaces of three systems: The flying vehicle must independently complete the aircraft model certification; the production system must meet the requirements for continued airworthiness; the operation end also needs to solve pilot qualifications, airspace, takeoff and landing facilities, and service networks. If any link is not closed, even if the land vehicle achieves large-scale mass production, it cannot be transformed into legal high-frequency operation of the flying vehicle. Therefore, publicly disclosed pre-sales and overseas orders are more suitable to be regarded as market signals rather than delivery evidence.
V. Mission Division of Labor for Four Technical Routes
Reclassifying the above models by configuration, we will get a clear "mission division of labor map."
At this stage, no configuration can cover all missions. Multi-rotors are suitable for short-range, controllable takeoff and landing sites, and standardized mission scenarios; compound-wings are suitable for cargo, offshore operation and maintenance, emergency rescue, and regional passenger transport; tilt-rotors are more suitable for intercity missions sensitive to speed and range; detachable flying cars attempt to connect ground travel and personal experience.
Therefore, it is invalid to judge the advantages and disadvantages of configurations using a single indicator, such as range, maximum takeoff weight, or number of seats. A truly effective evaluation framework should at least consider simultaneously: effective payload, range, noise, redundant safety, airworthiness difficulty, cost per mission, and the real density of mission scenarios.
Summary: The four routes have no advantages or disadvantages, only suitability. But there is a common rule worth remembering: The more a configuration pursues the performance ceiling, the longer the distance from "being able to fly" to "being able to be certified." The reason compound-wing is the most crowded is not because it is the best, but because it can be proven safe the fastest. This rule should directly enter the route selection decision of every enterprise.
Chapter 3 Rules: Airworthiness is the Real Watershed of This Industry
Up to here, all eVTOLs can be placed on the same progress map. Airworthiness status distinguishes enterprise levels better than release time, financing scale, or order quantity.
Figure 5: Green is certified, blue is airworthiness accepted/under certification, orange is engineering verification node, purple is debut appearance. The bar length only indicates the verifiable public progress interval and does not represent the actual construction period.
I. What the Three Certificates Manage Respectively
Many reports vaguely write "obtaining TC" as "getting the airworthiness certificate," which is inaccurate. The differences among the three are crucial:
TC (Type Certificate): Proves that the design of this model meets airworthiness standards. It is a design-level approval, unrelated to a specific aircraft.
PC (Production Certificate): Proves that the enterprise's production system can continuously and stably manufacture aircraft that meet the approved design. It manages the "quality system," not "individual products."
AC (Airworthiness Certificate): Proves that the specific aircraft in front of you is in a safe and usable condition, and can be registered and flown.
Only when all three are available simultaneously is the qualification foundation for "mass production + sales + operation" established. Having only TC without PC means the design is approved but cannot be mass-produced yet; having only a maiden flight without TC means it can fly but cannot legally carry passengers yet.
The four technical routes are essentially four different compromise methods for these three constraints.
II. Overview of Airworthiness and Commercialization Status
III. A Rule Worth Noting
The two models that completed the closed loop both avoided the most complex path of "manned passenger transport." EH216-S is an unmanned passenger-carrying aircraft, and V2000CG is a ton-class unmanned cargo aircraft. They still need to meet the complete aircraft airworthiness and production system requirements, but in terms of cockpit man-machine interfaces, passenger protection, pilot responsibilities, and operation rules, the mission boundaries are much narrower than those of manned passenger transport.
This means: The success of EH216-S and V2000CG cannot be directly extrapolated to mean "manned air taxis are mature." The competition focus of passenger eVTOL is shifting from "whether it can take off" to "whether a complete set of airworthiness archives can be established." And publicly available materials usually only disclose "accepted" or "entered compliance verification," rarely fully disclosing the certification basis, compliance matrix, and test flight conclusions. Therefore, at this stage, the certification date cannot be deduced based on enterprise plans, and "entering review" certainly cannot be written as "about to be certified."
A verifiable time reference: VE25-100 and E20 were accepted in September and October 2023 respectively. As of September 2026, it has been three years, and they are still in the compliance verification stage. The "acceptance—certification" cycle is measured in years, not quarters. This number should directly enter the cash planning of every enterprise.
But "complete three certificates" does not mean "already profitable." Whether sustainable cash flow can be formed depends on route density, payload utilization rate, maintenance system, airspace coordination capability, and the real cost per sortie. Airworthiness is the entry ticket, not the income statement.
IV. Orders and Deliveries: The Fog in the Numbers
In news reports on the low-altitude economy, order numbers are the most communicative part, and also the most easily misread part. The currently publicly visible main numbers are: Volante VE25-100 disclosed 260 aircraft (60 from Yuntian Group Indonesia, 200 from Minsheng Financial Leasing); ZG-ONE has thousand-level intended orders; AR-E800 had 160 confirmed and intended orders at the Tianjin Helicopter Expo; XPENG AEROHT Land Aircraft Carrier has pre-sale and overseas order disclosures; AE200 has the enterprise's claim of "hundreds of aircraft."
These numbers must be understood in layers:
Therefore, the correct reading is: Order quantity is more suitable for observing channel capability and market attention, rather than verifying product maturity. Public information generally does not disclose the deposit ratio, cancellation terms, delivery nodes, and airworthiness prerequisites. Adding them up to say "China's eVTOL has obtained thousands of orders" is misleading.
Looking at real deliveries in reverse: EH216-S completed the first batch of deliveries in 2023, and V2000CG achieved single aircraft delivery in July 2025. The common point of these two cases is—both happened after certification. This sequence is not a coincidence, but an industry rule.
Summary: Airworthiness is the only hard currency in this industry. It determines three things: whether it can be legally sold, whether it can be mass-produced, and whether it can fly continuously. And "orders" are soft signals, and "maiden flights" are intermediate nodes. Getting the priority of these three wrong is the most common strategic misjudgment at present.
Chapter 4 Landscape: Who is Building, Who Might Win
Figure 6: Among the 15 publicly unveiled eVTOL projects, the number of startup company models accounts for the absolute majority, but the three types of entities must ultimately accept the same set of aviation product verification and airworthiness constraints.
I. Endowments and Shortcomings of Three Types of Players
Startup Companies: Quantity Advantage, but Also the Greatest Pressure
12 models from AutoFlight, EHang, Volante, TIME, Lanyi, UFUTURE, Zero Gravity, Blue Vector, Muyu Tian, and others come from startup companies. Their advantages are high organizational efficiency and fast product iteration, enabling rapid trial and error in segmented tracks such as cargo, cultural tourism, medical care, and regional commuting. AutoFlight advancing on both cargo and passenger lines simultaneously, and EHang maintaining both EH216-S and VT35 simultaneously, rely precisely on this flexibility.
The shortcomings are equally obvious: large capital consumption, high test flight risks, long airworthiness cycles, and the supply chain and continued airworthiness system require long-term investment. The "number of debuts" of startup companies does not represent the final industrial concentration. As airworthiness and production consistency requirements deepen, resources will flow more to enterprises that can simultaneously complete engineering verification, quality systems, and operation organizations.
Automotive and Tech Cross-Boundary: Strong Mass Production Capability, but Cannot Replace Certification
AEROFUGA relies on the Geely Technology ecosystem, XPENG AEROHT relies on the XPENG ecosystem, and the GAC system is also continuously investing in low-altitude mobility. Such enterprises can more quickly reuse three-electric technology, intelligent manufacturing, supply chain management, and consumer operation capabilities—this is precisely the capability most needed when eVTOL goes from "building one" to "building one hundred."
But automotive mass production experience cannot replace aviation product certification. The recall logic and quality system of the automotive industry are not the same as the continued airworthiness, single aircraft release, and maintenance record system of the aviation industry. The acceptance of XPENG AEROHT's PC application precisely shows that cross-boundary enterprises also need to establish an aviation-grade production system from scratch.
Aviation Industry System: Not Much Exposure, but the Most Solid Engineering Capability
AVIC's AR-E800 is currently the only included model from the traditional aviation system. It emphasizes modular cargo compartments, hoisting capabilities, and complex terrain transportation, reflecting the reliance of heavy-load eVTOL on engineering reliability, mission adaptability, and aviation manufacturing systems. The early advantage of this route is not in consumer exposure, but in mission payload, operating environment, and engineering verification.
The three types of entities have different endowments, but the same endpoint: they all need to submit the same set of compliance evidence. This is also why "whose background is stronger" is a secondary question, and "whose certification progress is more solid" is the primary question.
II. Path Selection: Understanding the "Easy First, Hard Later" Certification Strategy in One Picture
Combining the three variables of "whether carrying passengers," "whether manned," and "openness of operation scenarios," we can obtain a four-level certification difficulty ladder. This ladder has direct tactical value for any enterprise planning the certification sequence.
Figure 7: The classification basis is "whether carrying passengers × whether manned × openness of operation scenarios," which is an illustrative classification rather than a quantitative score
First, what domestic eVTOL has verified are the two easiest levels. Level 1 (unmanned cargo) has V2000CG, and Level 2 (unmanned passenger) has EH216-S. Level 3 (manned, designated missions) is almost a blank zone—this is actually an intermediate market that has not yet been fully cultivated, which may be more realistic than directly impacting Level 4. Level 4 (manned, open passenger transport) gathers almost all mainstream projects under research, including AE200, E20, LE200, VE25-100, ZG-T6, Shengshilong, M1 passenger type, and Land Aircraft Carrier, all of which have not yet been certified.
Second, "cargo first, passenger later; unmanned first, manned later" is a proven feasible path. AutoFlight uses the V2000 series to layout both cargo and passenger simultaneously. Cargo is certified first, and passenger follows, accumulating platform data and supply chain, and using cargo missions to support the team and cash flow. This strategy of "exchanging time and data for lower difficulty levels" is worth serious evaluation by all enterprises.
III. An Easily Ignored Intermediate Zone: Level 3
It is worth pointing out separately that the level of "manned + designated mission operation" currently has almost no explicitly publicly disclosed projects under research in China, while internationally, scenarios such as agricultural and forestry operations, power line inspection, medical transfer, and sea pilotage—"manned but mission closed"—are often the realistic starting point for eVTOL commercialization. Its value lies in: it can not only avoid the public acceptance and remote control reliability difficulties of unmanned driving, but also does not have to face the high-density operation rules under open passenger transport scenarios.
For enterprises that have not yet locked in their routes, this is a strategic gap worth evaluating: Instead of competing head-on in the already crowded Level 4, it is better to establish certification and operation experience in Level 3 first.
Summary: The core conclusion of the industrial landscape: Background determines the starting point, path determines the speed, and system determines the endpoint. Startup companies have a quantity advantage but limited ammunition, cross-boundary enterprises have money and supply chains but need to make up for aviation system lessons, and the aviation industry has engineering capabilities but lacks consumer scenarios. The ultimate winner will definitely be the one that first figures out the certification sequence and allocates resources accordingly.
Chapter 5 Landscape: What to Watch, What to Do
If only one judgment is used to summarize the current stage, it is: Domestic eVTOL has collectively crossed the "prototype competition" and entered the "airworthiness competition." The characteristic of this stage is that press conferences no longer produce decisive impacts; the real watershed appears in regulatory documents and delivery records.
I. Six Key Observation Indicators
For models that have entered certification: AE200, LE200, VE25-100, E20, Shengshilong, M1, ZG-T6, VT35, ZG-ONE, and Land Aircraft Carrier X3-F, the six most worth tracking pieces of public information are:
1. Whether all compliance verification is completed, which is the last technical hurdle before certification;
2. Whether TC is obtained, and the applicable PC. The former is design approval, and the latter is mass production approval;
3. Whether single aircraft AC is issued to form verifiable delivery records, which is the final kick from "product" to "commodity";
4. Whether the production system, quality records, and maintenance system are disclosed. This is the core of PC and also where the long-term competitiveness of the enterprise lies;
5. Whether a stable and auditable delivery rhythm is formed. One-time delivery does not constitute commercialization;
6. Whether real mission data and continuous operation data are accumulated. This is the only evidence of operation economy.
And the establishment of large-scale requires the simultaneous availability of three types of conditions: product, operation, and infrastructure. On the product end, there must be complete airworthiness and stable mass production; on the operation end, there must be clear operation rules, pilot or remote driving systems, and airspace coordination mechanisms; on the infrastructure end, there must be takeoff and landing sites, charging or battery swapping networks, and maintenance and dispatch systems. If there is only aircraft certification but lacking missions and facilities, large-scale is still difficult to establish; if there are only orders and operation plans but lacking airworthiness products, stable delivery cannot be formed.
But automotive mass production experience cannot replace aviation product certification. The recall logic and quality system of the automotive industry are not the same as the continued airworthiness, single aircraft release, and maintenance record system of the aviation industry. The acceptance of XPENG AEROHT's PC application precisely shows that cross-boundary enterprises also need to establish an aviation-grade production system from scratch.
II. Nine Suggestions for OEMs
The following suggestions are all derived from the facts in the preceding text, and each corresponds to an industry case that has been verified or falsified.
Suggestion 1: Put "certification path design" before configuration design.
The common point of the two closed-loop cases is that they both avoided the most difficult combination of "manned + open airspace + large-scale passenger carrying." Before freezing the configuration, enterprises should first answer three questions: cargo first then passenger, or direct passenger? Unmanned first or manned? Is the first certified configuration the "minimum viable version"? If this sequence is wrong, all subsequent work will have to be reworked.
Suggestion 2: Platformization can reuse costs, but don't expect "one certificate for multiple uses."
AutoFlight's V2000CG and V2000EM share a platform, but the passenger model still needs to independently go through the complete certification. The inspiration is that platform reuse saves R&D, test flight data, supply chain, and quality system costs, but it cannot save the passenger-exclusive passenger protection, system redundancy, and cabin environment verification. Budgets and cycles should be arranged according to "independent certification," not according to "derivative models."
Suggestion 3: Freeze the certification basis as soon as possible. Repeated configuration is the biggest cycle killer.
E20 is still in the G-1, G-2, and compliance plan stages; the parameter calibers of some models in different batches are different, which itself reflects the traces of configuration iteration. It is recommended that enterprises draw a clear timeline between "stopping design iteration" and "maintaining technical advancement"—once the certification basis is repeated, the previously accumulated compliance evidence may all be invalidated.
Suggestion 4: Order disclosure should be categorized by caliber. Don't let numbers backfire.
260 aircraft (including leasing and intentions), thousand-level intentions, 160 confirmed + intentions—once these three sets of numbers are merged by the outside world into "thousands of orders," it will form a wrong expectation baseline. It is recommended to uniformly adopt the four-layer caliber of "intention/framework/confirmed/delivery" for separate disclosure, and proactively explain the airworthiness prerequisites. When delivery is delayed, transparent caliber is the best credit protection.
Suggestion 5: The supply chain should be managed as "airworthiness parts" rather than "automotive-grade parts" or "industrial parts."
The three-electric and supply chain advantages of cross-boundary enterprises truly exist, but aircraft require traceability to individual parts, possession of qualification certificates, and provision of continued airworthiness support. The acceptance of XPENG AEROHT's PC application shows that this step cannot be bypassed. It is recommended to establish material traceability and a qualified supplier list starting from the first prototype, rather than making up accounts when applying for PC.
Suggestion 6: PC should be laid out 12-18 months earlier than the first aircraft delivery.
PC manages the production system, not individual products. Starting to build a production line only after getting TC will often waste more than a year. The experience of EH216-S and V2000CG is that certification and the production system are promoted simultaneously. Treat PC as the second battlefield parallel to TC, rather than the finishing work after certification.
Suggestion 7: Scenario selection follows "closed first, open later; special first, mass later."
EH216-S falls in cultural tourism and sightseeing, V2000CG falls in islands, offshore platforms, and emergencies, and M1 cuts into medical rescue. The common point is: clear mission boundaries, coordinable airspace, relatively insensitive to cost, or high social value. Such scenarios are not "settling for second best," but necessary springboards for accumulating operation data, integrating operation systems, and cultivating public acceptance.
Suggestion 8: The pace of funding should match the certification cycle, not the financing rhythm.
A verifiable reference: VE25-100 and E20 were accepted in September-October 2023, and as of September 2026, they are still in compliance verification, having reached three years. It is recommended that enterprises arrange cash reserves and financing rhythms according to "at least 3-4 years after acceptance," treating "certification" as a long-term project that requires continuous investment, rather than a milestone that can be sprinted.
Suggestion 9: Make up for three shortcomings in advance organizationally—quality system, continued airworthiness, and operation support.
Startup companies are generally strong in design but weak in systems. And PC, single aircraft release, maintenance records, and continued airworthiness documents are precisely the keys to whether continuous operation can be achieved after certification. It is recommended to start building an independent quality and airworthiness department 24 months before obtaining TC, reporting directly to the top management.
III. Suggestions for Other Roles
To cross-boundary enterprises: Do not treat aircraft as smart hardware for iteration. The "release first, fix later with OTA" approach in the automotive field means the invalidation of the certification basis and evidence in the aviation field. A feasible approach is to limit rapid iteration to ground systems, cockpit experience, and operation software, and insist on freezing flight-critical systems.
To supply chain enterprises: The real opportunity brought by eVTOL is not in the complete aircraft, but in "airworthiness-grade" three-electric, composite materials, flight control, and actuation systems. Whoever can first provide domestic parts with qualification certificates, traceability, and support for continued airworthiness will be able to lock in their position when this round of industrialization takes shape. The current period is precisely the window for domestic substitution.
To operators and local governments: The number of certified models is increasing, but "places where they can fly" are still the bottleneck. It is recommended to prioritize investment in takeoff and landing site networks, charging/battery swapping facilities, airspace coordination mechanisms, and maintenance capabilities. Without infrastructure, certified models have nowhere to fly; without mission scenarios, infrastructure has no way to recover investment—the planning rhythms of the two need to be aligned.
To investors: Shift the focus of due diligence away from "parameter leadership" and "order scale" to four harder-to-fake and more critical indicators: whether the certification basis is frozen, the completion degree of compliance verification, the readiness of the quality system and PC, and the matching degree of cash consumption rate and certification cycle. In this industry, the most expensive thing is not technology, but time.
IV. Six Pitfalls to Avoid
Conclusion: Between "Being Able to Fly" and "Being Able to Use" Lies Patience
Looking back at the hustle and bustle of these three years, there are actually only two dates worth remembering: October 2023, when EH216-S obtained the Type Certificate; March 2024, when V2000CG obtained the Type Certificate. Everything that happened after these two time points: deliveries, commercial demonstrations, single aircraft airworthiness certificates, and production certificates, are all derived from these two pieces of paper. And everything that happened before these two time points, debuts, maiden flights, orders, and financing, have not yet undergone final regulatory confirmation.
For the enterprises in it, the core proposition for the next two or three years is actually very simple:
First, choose the right path. Easy first, hard later is not a compromise, but a proven certification strategy—the two certified models are both at the bottom two levels of the difficulty ladder.
Second, freeze the configuration. Finalize the design before the certification basis is repeated, because compliance evidence is the most expensive sunk cost.
Third, build a good system. Quality system, continued airworthiness, and operation support—these "unsexy" capabilities determine whether the aircraft can really be delivered and kept flying after certification.
Fourth, find the right scenarios. Closed first, open later; special first, mass later. Use replicable missions to polish the operation model, rather than waiting for a "mass air travel market" that does not yet exist.
Fifth, manage time well. The most expensive cost in this industry is time, and the longest part of time is called "compliance verification." Financing rhythm, organizational patience, and investor expectations must be aligned by the year, not by the quarter.
As of the publication date of this article, no publicly verifiable three-certificate closed-loop model of the same level as V2000CG has yet appeared on the passenger transport end. This state is not industrial stagnation, but a normal stage of the industry shifting from "prototype competition" to "airworthiness competition." What is truly worth waiting for in the next two years is not the release of another concept aircraft, but a manned passenger eVTOL completing the entire TC-PC-AC process and stably delivering the first batch of paying passengers to their destinations.
When that day comes, the low-altitude economy will truly begin.