Truck and unmanned aerial vehicle combined distribution path planning method and system suitable for complex environment

Abstract

The invention relates to the technical field of logistics distribution, and discloses a truck and unmanned aerial vehicle combined distribution path planning method and system suitable for a complex environment, and the method comprises a multi-topographic feature fusion module which comprises data collection and data processing; the data processing module analyzes the current terrain according to a multi-terrain feature fusion algorithm, the multi-terrain feature fusion module is connected with a path planning module, and the path planning module plans corresponding paths by using different algorithms for different terrains according to a judgment result of the multi-terrain feature fusion module; and the path planning module is connected with a feedback display module, and the feedback display module displays the planned path. According to the method, the influence of complex terrains such as mountainous regions, water areas and jungles on the operation of the truck and the unmanned aerial vehicle is comprehensively considered, and scientific planning and dynamic adjustment of the path are realized by applying an advanced algorithm and a real-time monitoring technology.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics distribution, and in particular to a method and system for joint distribution path planning of trucks and drones adapted to complex environments. Background Art

[0002] With the rapid development of the global economy and the booming rise of e-commerce, the logistics industry has faced unprecedented opportunities and challenges. Consumers' requirements for the timeliness and service quality of commodity distribution are increasing day by day. They not only expect the goods to be delivered quickly but also require the distribution process to be more accurate and reliable. Against this background, the logistics distribution demand shows a trend of diversification and personalization. The number of small-batch and multi-batch distribution orders is gradually increasing, and the distribution scope is constantly expanding, covering all corners of the city and remote rural areas. When facing such complex and changeable demands, the traditional logistics distribution system has begun to expose many limitations.

[0003] Traditional logistics distribution mainly relies on ground transportation tools such as trucks. Although trucks have a large load capacity and relatively stable transportation capabilities and can meet the distribution needs of most conventional goods, in actual operation, they face many problems. In cities, the traffic congestion problem is becoming increasingly serious. Especially during peak hours, the road traffic efficiency is greatly reduced, and the truck distribution time is significantly extended. This not only increases the logistics cost but also seriously affects the customer experience. For remote areas or regions with inconvenient transportation, due to the imperfect road infrastructure, truck transportation often faces problems such as difficult passage and long transportation time, making it difficult to achieve efficient distribution. In order to break through the bottleneck of traditional distribution methods, the joint distribution mode of trucks and drones has emerged. Drones, with their advantages of being unrestricted by ground traffic, strong mobility, and high speed, can play an important role in short-distance distribution. Especially in the last-mile distribution in cities and emergency distribution in remote areas, they have great application potential. Combining the large-capacity and long-distance transportation capabilities of trucks with the flexibility of drones can achieve complementary advantages and is expected to bring new solutions for logistics distribution.

[0004] However, in actual applications, the joint distribution of trucks and drones faces complex and changeable environmental challenges. Different topographies and landforms, such as mountains, waters, jungles, etc., bring great difficulties to the distribution path planning. In mountainous areas, factors such as slope and wind speed will affect the driving speed of trucks and the flight stability of drones; problems such as water surface reflection and water flow in water environments will interfere with the positioning and flight safety of drones; the dense vegetation, complex terrain, and signal occlusion in jungle areas also increase the difficulty of distribution. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a method and system for joint distribution path planning of trucks and drones adapted to complex environments. Different calculation formulas are used for mountains, waters, and jungles respectively, solving the problem that in practical applications, the joint distribution of trucks and drones faces complex and changeable environmental challenges. Different topographies (mountains, waters, jungles, etc.) affect the driving of trucks and the flight of drones respectively from aspects such as slope, wind speed, water surface reflection, vegetation topography, and signal occlusion. In severe cases, it will lead to the problem that the distribution cannot be delivered.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A joint distribution path planning system for trucks and drones adapted to complex environments, including a multi-topography feature fusion module, and the multi-topography feature fusion module includes data collection and data processing; the data processing analyzes the current topography according to the multi-topography feature fusion algorithm. A path planning module is connected to the multi-topography feature fusion module. The path planning module uses different algorithms to plan corresponding paths for different topographies according to the judgment results of the multi-topography feature fusion module; a feedback display module is connected to the path planning module, and the feedback display module displays the planned path.

[0007] Preferably, the data collection in the multi-topography feature fusion module includes: on-vehicle slope sensors, satellite remote sensing and hydrological monitoring equipment, and on-vehicle sensors.

[0008] Preferably, the data processing algorithms in the multi-topography feature fusion module include:

[0009] I. Definition of terrain membership function

[0010] Input parameter and formula symbol table

[0011] The variable symbols include: S norm (Normalized slope (angle)), D forest (Jungle density index), P water (Water area probability) and δ (classification judgment buffer threshold);

[0012] Among them, S norm (Normalized slope (angle)) is obtained using on-vehicle slope sensors; D forest (Jungle density index) is obtained by multiplying the infrared reflectivity by the number of trees per unit area; P water (Water area probability) is obtained by multiplying the water area reflectivity value by the real-time soil humidity; δ (classification judgment buffer threshold) represents a fixed empirical value;

[0013] Membership function

[0014] 1. Mountain membership:

[0015]

[0016] Physical meaning: The higher the slope and the lower the jungle density, the higher the probability of mountainous terrain;

[0017] Inhibiting term: In the denominator Suppress the possibility of misjudging mountainous terrain in the jungle;

[0018] 2. Jungle membership:

[0019]

[0020] Physical meaning: Nonlinear amplification of the jungle density index (exponent 1.2), logical gate Suppress misjudgment of the jungle when the slope > 12°;

[0021] 3. Water area membership:

[0022]

[0023] Physical meaning: The probability of the water area is strongly correlated with the terrain flatness (quadratic term eliminates misjudgment of the water area when the slope > 10°).

[0024] Preferably, the data processing algorithm in the multi-terrain feature fusion module further includes:

[0025] II. Decision rules and classification logic

[0026] Step 1: Calculate membership

[0027] For the slope, jungle density, and water area probability data collected in real time, calculate the membership values M mountain , M forest , M water ;

[0028] Step 2: Threshold determination

[0029] Execute the following conditional logic (priority order):

[0030] 1. Mountain priority rule:

[0031]

[0032] Logical extension: When the slope is dominant (such as in steep slope areas) and the buffer threshold δ = 0.15 is satisfied, it is still judged as mountainous even if the jungle density is high;

[0033] 2. Jungle mandatory condition:

[0034]

[0035] Safety limit: Only when the jungle density index D forest ≥ 1.2 does it take effect;

[0036] 3. Absolute determination of water area:

[0037]

[0038] Dual - insurance mechanism: Directly with high confidence (M water > 0.7); or with medium membership degree but extremely high original water - area probability (P water > 0.8).

[0039] Preferably, the path - planning module includes: a mountain - terrain planning module, a water - area terrain planning module, and a jungle - terrain planning module.

[0040] Preferably, in the mountain - terrain planning module, the detour path of the route and the altitude of the drone are adjusted based on various parameters of the mountain and equipment parameters. The specific algorithm is as follows:

[0041] Input:

[0042] 1. Parameters: S current represents the current slope; C rough : represents the ruggedness; D critical represents the terrain no - entry mark;

[0043] 2. Cost function:

[0044]

[0045] Coefficient description:

[0046] α = 0.6: Path - length weight

[0047] β = 10 6 : Penalty term for forced detour around steep slopes

[0048] γ = 0.4: Ruggedness smoothing coefficient (C max = 5g 2 / Hz)

[0049] 3. Drone altitude adjustment rule:

[0050] H uav = H base + 0.3·Δh peak (Δh peak is the local mountain - peak height difference)

[0051] Output: Truck detour path set + Drone elevation trajectory.

[0052] Preferably, in the water - area terrain planning module, the water - avoiding path of the route and the drone's cross - water flight path are adjusted based on various parameters of the water area and equipment parameters. The specific algorithm is as follows:

[0053] Input:

[0054] 1. Water area mask generation: If P water > 0.5, mark it as a no-go area; Expand the safety boundary: L safe = max(10m, 0.1·D moist )(humidity-related buffer)

[0055] 2. Truck path cost: v road : Road type speed table (e.g., for dirt road, v = 20 km / h)

[0056] 3. UAV flight path correction: Humidity reflection compensation height: H comp = H nominal + 0.05·ln(1 + D moist ); Bypass logic: If the water area width > 50m, force phased crossing (each section s 30m)

[0057] Output: Truck water avoidance path + UAV water crossing flight path.

[0058] Preferably, in the jungle terrain planning module, the truck jungle passage path and the UAV dynamic obstacle avoidance path are allocated based on various parameters of the jungle and equipment parameters. The specific algorithm is as follows:

[0059] Input:

[0060] 1. Jungle density grid: Unit grid 5m × 5m, density threshold:

[0061]

[0062] D pass = 1, it is determined that trucks are allowed to pass; D pass = 2.5, it is determined that only UAVs can pass;

[0063] 2. Truck path penalty term:

[0064]

[0065] 3. UAV obstacle avoidance logic:

[0066] Real-time scanning radius R scan = 15m

[0067] Obstacle avoidance priority: If the obstacle distance < 5m Vertical climb ΔH = 3m + 0.1v uav

[0068] Output:

[0069] Truck jungle passage + UAV dynamic obstacle avoidance path.

[0070] Usage method of a combined distribution path planning system for trucks and drones adapted to complex environments, including the following steps:

[0071] S1. Data collection: Real-time obtain slope, jungle density, and water area probability data through on-vehicle sensors (slope / vibration), satellite remote sensing (accuracy ≤ 1 m), and hydrological equipment (sampling frequency ≥ 10 Hz). Use drones for assisted high-precision positioning (RTK, ≤ 2 cm). Then calculate the terrain feature weights using the membership functions of mountains, jungles, and water areas, and determine the main terrain type according to the priority rules (steep slopes first, high confidence in water areas, etc.), and mark the no-go areas;

[0072] S2. Data analysis: And according to the results of data collection and judgment, including mountains, water areas, and jungles, use corresponding algorithms for path planning and analysis:

[0073] S3. Display and feedback: Display the obtained path planning through a display device.

[0074] The present invention provides a collaborative distribution path planning system for trucks and heterogeneous drones based on dynamic requirements.

[0075] Has the following beneficial effects:

[0076] 1. The present invention is provided with a path planning module, and according to different scenarios, such as mountains, water areas, and jungles, respectively according to the characteristics of their terrain, such as steep slopes, water accumulation areas, and tree obstacles, use different algorithms for route planning, further ensuring the accuracy of path planning.

[0077] 2. The present invention adopts a multi-source data fusion technology of satellite remote sensing (accuracy ≤ 1 m), hydrological monitoring (sampling frequency ≥ 10 Hz), and on-vehicle slope sensors, and introduces a priority classification rule to intelligently distinguish mountains, water areas, and jungles, making the judgment more accurate during the process of path planning classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 Is the system flow chart of the present invention;

[0079] Figure 2 Is the method flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0081] Embodiment:

[0082] Please refer to the attached Figure 1 - Attachment Figure 2 , the embodiment of the present invention provides a combined distribution path planning system for trucks and drones adapted to complex environments, including a multi-terrain feature fusion module, and the multi-terrain feature fusion module includes data acquisition and data processing;

[0083] The data acquisition uses on-vehicle slope sensors, satellite remote sensing and hydrological monitoring equipment, and on-vehicle sensors to collect terrain data in different ways for different terrains to ensure the accuracy of terrain data. Moreover, the satellite remote sensing data accuracy ≤ 1 m (such as Landsat-9), and the sampling frequency of the hydrological monitoring equipment ≥ 10 Hz; the drone positioning system uses RTK technology, and the horizontal positioning error ≤ 2 cm;

[0084] The data processing analyzes the current terrain according to the multi-terrain feature fusion algorithm, and the specific algorithm is as follows:

[0085] I. Definition of terrain membership function

[0086] Input parameter and formula symbol table

[0087] Variable symbols include: S norm (Normalized slope (angle)), D forest (Jungle density index), P water (Water area probability) and δ (classification determination buffer threshold);

[0088] Among them, S norm (Normalized slope (angle)) is obtained using an on-vehicle slope sensor; D forest (Jungle density index) is obtained by multiplying the infrared reflectance by the number of trees per unit area; P water (Water area probability) is obtained by multiplying the water area reflectance value by the real-time soil humidity; δ (classification determination buffer threshold) represents a fixed empirical value;

[0089] Membership function

[0090] 1. Mountain membership:

[0091]

[0092] Physical meaning: The higher the slope and the lower the jungle density, the higher the probability of mountains.

[0093] Inhibitory term: In the denominator Suppress the possibility of misjudging mountains in the jungle.

[0094] 2. Jungle membership:

[0095]

[0096] Physical meaning: Nonlinear amplification of jungle density index (exponent 1.2), logic gate Suppress jungle misjudgment when the slope > 12°.

[0097] 3. Water membership degree:

[0098]

[0099] Physical meaning: The water probability is strongly correlated with terrain flatness (quadratic term eliminates water misjudgment when the slope > 10°).

[0100] II. Judgment rules and classification logic

[0101] Step 1: Membership degree calculation

[0102] Calculate three types of membership degree values M mountain , M forest , M water .

[0103] Step 2: Threshold judgment

[0104] Execute the following conditional logic (priority order):

[0105] 1. Mountain priority rule:

[0106]

[0107] Logical extension: When the slope is dominant (such as in steep slope areas) and the buffer threshold δ = 0.15 is satisfied, even if the jungle density is high, it is still judged as mountainous.

[0108] 2. Jungle mandatory condition:

[0109]

[0110] Safety limit: Only when the jungle density index D forest ≥ 1.2 does it take effect (to avoid misjudgment of sparse vegetation).

[0111] 3. Water absolute judgment:

[0112]

[0113] Double insurance mechanism: Directly with high confidence (M water > 0.7); or medium membership degree but extremely high original water probability (P water > 0.8).

[0114] Algorithm judgment example:

[0115] Scenario 1: When M mountain = 0.55, Mforest = 0.3, M water = 0.2, then the result: 0.55 > max(0.3, 0.2) + 0.15 → 0.55 > 0.45, judged as mountainous area;

[0116] Scenario 2: When M forest = 0.6, M mountain = 0.5, M water = 0.3, and D forest = 2.0, then the result: 0.6 > max(0.5, 0.3) + 0.15 → 0.6 > 0.657 is not satisfied, but the mandatory condition D forest ≥ 1.2 is triggered, judged as jungle;

[0117] Scenario 3: When M water = 0.65, M mountain = 0.5, M forest = 0.4, then the result: directly satisfy M water > 0.7? Not satisfied, but if P water = 0.85, then the double - insurance condition is triggered, judged as water area.

[0118] A path planning module is connected to the multi - terrain feature fusion module. The path planning module uses different algorithms to plan corresponding paths for different terrains according to the results judged by the multi - terrain feature fusion module; A feedback display module is connected to the path planning module. The feedback display module displays the planned path to facilitate the staff to view and understand.

[0119] The path planning module includes: mountain terrain planning module, water area terrain planning module, jungle terrain planning module;

[0120] In the mountain terrain planning module, the detour path of the route and the altitude of the UAV are adjusted based on the parameters of the mountain and the equipment parameters. The specific algorithm is:

[0121] Input:

[0122] 1. Parameters: S current represents the current slope; C rough : represents the ruggedness (measured by the vehicle - mounted vibration sensor); D critical represents the terrain no - entry mark;

[0123] 2. Cost function:

[0124]

[0125] Coefficient description:

[0126] α = 0.6: path length weight

[0127] β = 10 6 : Penalty term for forced detour around steep slopes

[0128] γ = 0.4: Roughness smoothing coefficient (C max = 5g 2 / Hz)

[0129] 3. UAV altitude adjustment rule:

[0130] H uav = H base + 0.3·Δh peak (Δh peak is the elevation difference of local peaks)

[0131] Output: Truck detour path set + UAV elevation trajectory.

[0132] In the water area terrain planning module, the water avoidance path of the route and the UAV cross - water flight path are allocated based on various parameters of the water area and equipment parameters. The specific algorithm is as follows:

[0133] Input:

[0134] 1. Water area mask generation: If P water > 0.5, mark it as a no - go area; Expand the safety boundary: L safe = max(10m, 0.1·D moist )(humidity - related buffer)

[0135] 2. Truck path cost: v road : Road type speed table (such as for dirt road, v = 20km / h)

[0136] 3. UAV flight path correction: Humidity reflection compensation height: H comp = H nominal + 0.05·ln(1 + D moist ); Bypass logic: If the water area width > 50m, force phased crossing (each section s 30m)

[0137] Output: Truck water avoidance path + UAV cross - water flight path.

[0138] In the jungle terrain planning module, the truck jungle access path and the UAV dynamic obstacle avoidance path are allocated based on various parameters of the jungle and equipment parameters. The specific algorithm is as follows:

[0139] Input:

[0140] 1. Jungle density rasterization: Unit grid 5m×5m, density threshold:

[0141]

[0142] D pass If D = 1, it is determined that trucks are allowed to pass; D pass If D = 2.5, it is determined that only drones can pass; 2. Truck path penalty term:

[0143]

[0144] 3. Drone obstacle avoidance logic:

[0145] The real-time scanning radius R scan = 15m

[0146] Obstacle avoidance priority: If the obstacle distance < 5m Vertical climb ΔH = 3m + 0.1v uav Output:

[0147] Truck jungle passage + Drone dynamic obstacle avoidance path.

[0148] Full table of path planning algorithm symbols:

[0149]

[0150]

[0151] The usage method of the combined distribution path planning system for trucks and drones adapting to complex environments includes the following steps:

[0152] S1. Data collection: Real-time obtain slope, jungle density and water area probability data through in-vehicle sensors (slope / vibration), satellite remote sensing (accuracy ≤ 1m) and hydrological equipment (sampling frequency ≥ 10Hz), and use drones for high-precision positioning (RTK, ≤ 2cm). Then calculate the terrain feature weights using the membership functions of mountains, jungles and water areas, and determine the main terrain type according to the priority rules (steep slopes first, high confidence in water areas, etc.), and mark the no-go areas;

[0153] S2. Data analysis: And according to the results of data collection and judgment, including mountains, water areas and jungles, use the corresponding algorithms for path planning and analysis. The use of the corresponding algorithms for path planning and analysis includes:

[0154] Mountain planning: Trucks avoid steep slopes (detour when the slope > 22.5°), and drones dynamically lift the height according to the local terrain (such as +30% of the peak-valley height difference); Water area planning: Trucks bypass the water accumulation area based on the humidity mask, and drones cross large water areas through segmented flight (each segment ≤ 30m) and humidity compensation height; Jungle planning: Trucks are only allowed to pass through low-density jungles (D forest < 2.5), and drones scan obstacles within a 5m radius in real time and avoid them vertically;

[0155] S3. Display and feedback: Display the obtained path planning through a display device to facilitate viewing and understanding by the staff.

[0156] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A truck and drone joint distribution path planning system adapted to complex environments, characterized in that, It includes a multi - terrain feature fusion module, and the multi - terrain feature fusion module includes data collection and data processing; the data processing analyzes the current terrain according to the multi - terrain feature fusion algorithm. A path planning module is connected to the multi - terrain feature fusion module. The path planning module uses different algorithms to plan corresponding paths for different terrains according to the judgment result of the multi - terrain feature fusion module; a feedback display module is connected to the path planning module, and the feedback display module displays the planned path.

2. The truck and drone joint distribution path planning system adapted to complex environments according to claim 1, wherein The data collection in the multi - terrain feature fusion module includes: vehicle - mounted slope sensors, satellite remote sensing and hydrological monitoring equipment, and vehicle - mounted sensors.

3. The truck and drone joint distribution path planning system adapted to complex environments according to claim 1, wherein The data processing algorithms in the multi - terrain feature fusion module include: I. Definition of terrain membership function Input parameter and formula symbol table The variable symbols include: S norm (Normalized slope (angle)), D forest (Jungle density index), P water (Water area probability) and δ (Classification determination buffer threshold); Among which S norm (Normalized slope (angle)) is obtained using an on-vehicle slope sensor; D forest (Jungle density index) is obtained by infrared reflectance × number of trees per unit area; P water (Water area probability) is obtained by water area reflectance value × real-time soil humidity; δ (classification determination buffer threshold) represents a fixed empirical value; Membership function 1. Mountain membership: Physical meaning: The higher the slope and the lower the jungle density, the higher the probability of mountain. Suppression term: in the denominator Suppress the possibility of misjudging mountains in the jungle; 2. Jungle membership: Physical meaning: Nonlinear amplification of the jungle density index (exponent 1.2), logic gate Suppress jungle misjudgment when the suppression slope > 12°; 3. Water area membership: Physical meaning: The probability of water area is strongly correlated with terrain flatness (quadratic term eliminates misjudgment of water area when slope > 10°).

4. The truck and drone joint distribution path planning system adapted to complex environments according to claim 1, wherein The data processing algorithms in the multi - terrain feature fusion module also include: II. Decision rules and classification logic Step 1: Membership calculation For the slope, jungle density, and water area probability data collected in real time, calculate the membership values M of three categories moun tai n ,Mf ores t,M wa t er ; Step 2: Threshold judgment Execute the following conditional logic (priority order):

1. Mountain priority rule: Logical extension: When slope is dominant (such as in steep slope areas) and satisfies the buffer threshold δ = 0.15, even if the jungle density is high, it is still judged as mountain.

2. Jungle mandatory condition: Safety restriction: Only when the jungle density index D forest ≥ 1.2 does it take effect; 3. Water area absolute judgment: Else IfM water > 0.7 or (M water > 0.5 and P water > 0.8) Water terrain Dual insurance mechanism: direct high confidence (M water > 0.7); or medium membership but extremely high probability of the original water area (P water > 0.8).

5. The joint distribution path planning system for trucks and drones adapted to complex environments according to claim 1, wherein The path planning module includes: mountain terrain planning module, water area terrain planning module, and jungle terrain planning module.

6. The truck and drone joint distribution path planning system adapted to complex environments according to claim 5, characterized in that In the mountain terrain planning module, the detour path of the route and the height of the drone are adjusted based on various parameters of the mountain and equipment parameters. The specific algorithm is: Input:

1. Parameter: S current represents the current slope; C rough : represents the ruggedness; D critical represents the terrain no-go mark; 2. Cost function: Coefficient description: α = 0.6: Path length weight β = 10 6 : Penalty term for forced detour around steep slopes γ = 0.4: Roughness smoothing coefficient (C max = 5g 2 / Hz) 3. Drone height adjustment rule: H uav = H base + 0.3·Δh peak (Δh peak is the local mountain height difference) Output: Truck detour path set + Drone elevation trajectory.

7. The truck and unmanned aerial vehicle joint distribution path planning system adapted to complex environments according to claim 5, wherein In the water area terrain planning module, the water - avoiding path of the route and the cross - water route of the drone are adjusted based on various parameters of the water area and equipment parameters. The specific algorithm is: Input:

1. Water area mask generation: If P water > 0.5, mark it as a no-go area; Expand the safety boundary: L safe = max(10 m, 0.1·D moist )(humidity-related buffer) 2. Truck path cost: v road : Road type speed table (e.g., for dirt road, v = 20 km / h) 3. UAV Route Correction: Humidity Reflection Compensation Altitude: H comp = H nominal + 0.05 · ln(1 + D moist ); Flying-around Logic: If the water area width > 50m, force phased crossing (each section s 30m) Output: Truck water - avoiding path + Drone cross - water route.

8. The truck and drone joint distribution path planning system adapted to complex environments according to claim 5, wherein, In the jungle terrain planning module, the jungle passage path of the truck and the dynamic obstacle - avoidance path of the drone are adjusted based on various parameters of the jungle and equipment parameters. The specific algorithm is: Input:

1. Jungle density rasterization: Unit grid 5m×5m, density threshold: D pass If D = 1, it is determined that trucks are allowed to pass; D pass If D = 2.5, it is determined that only drones can pass; 2. Truck path penalty term:

3. Drone obstacle - avoidance logic: Real-time scanning radius R scan = 15 m Obstacle avoidance priority: If the obstacle distance < 5m Vertical climb ΔH = 3m + 0.1v uav Output: Truck jungle passage + Drone dynamic obstacle - avoidance path.

9. The method of using the joint distribution path planning system for trucks and drones adapted to complex environments according to any one of claims 1-8, characterized in that, It includes the following steps: S1. Data collection: Real - time obtain slope, jungle density and water area probability data through vehicle - mounted sensors (slope / vibration), satellite remote sensing (accuracy ≤ 1m) and hydrological equipment (sampling frequency ≥ 10Hz), and use drone - assisted high - precision positioning (RTK, ≤ 2cm). Then calculate the terrain feature weights using the membership functions of mountains, jungles, and water areas, and determine the main terrain type according to the priority rules (steep slopes first, high confidence in water areas, etc.), and mark the no - go areas. S2. Data analysis: Based on the results of data collection and judgment, including mountains, waters, and jungles, corresponding algorithms are used for path planning and analysis: S3. Display and feedback: The obtained path planning is displayed through a display device.

Images