Parking point recommendation method

Abstract

According to the parking spot recommendation method provided by the invention, a dynamic adaptive path cost model and a decision weight distribution mechanism are constructed by fusing real-time rainfall, a road network risk coefficient, a vehicle state and multi-dimensional information of a parking lot, path planning can be dynamically optimized according to rainfall intensity, a high-waterflooding risk road section is preferentially avoided, and the parking spot recommendation efficiency is improved. The weights of decision factors such as safety, distance, scene adaptability and economical efficiency are automatically adjusted according to the weather severity, so that the recommendation strategy takes safety as a primary target under an extreme condition, meanwhile, by distinguishing high-speed and non-high-speed different scenes and accurately matching the parking lot types, the reasonability and practicability of the recommendation result are remarkably improved, and the recommendation efficiency is improved. And finally, through integration sorting, safety early warning and multi-mode man-machine interaction guidance, intelligent decision support is provided for a driver in an emergency, and the driving safety and the risk avoiding efficiency are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of driver assistance technology, and in particular to a parking spot recommendation method. Background Technology

[0002] Current in-vehicle navigation and driver assistance systems have not effectively addressed the practical needs of drivers for temporary safe parking and evacuation in extreme weather conditions. The main problems are as follows: lack of targeted recommendations: navigation systems lack the ability to identify and recommend temporary parking spots (such as commercial buildings and service areas); slow response: unable to quickly determine a safe parking plan after extreme weather occurs; poor scenario adaptability: recommendation strategies lack differentiation under highway / non-highway conditions and different vehicle speeds. Summary of the Invention

[0003] This invention provides a parking spot recommendation method, which aims to overcome the shortcomings of the prior art and effectively improve driving safety and risk avoidance efficiency.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A parking spot recommendation method includes: S1. Data collection steps: Obtain vehicle status information, real-time rainfall, candidate parking lot information within the search radius, and road network data; S2. Parking lot filtering step: Based on the candidate parking lot information, filter out parking lots with no vacancies or those whose distance exceeds a preset threshold to obtain a set of effective candidate parking lots. S3. Path planning step: For each of the valid candidate parking lots, calculate the optimal path from the vehicle's current location to the parking lot, and output the path segment sequence; S4. Safety assessment step: Based on the road segment sequence of the optimal route, calculate the weighted average flooding probability of the route; S5. Recommendation factor calculation steps: Calculate the recommendation factor for each of the effective candidate parking lots, wherein the recommendation factor includes at least safety, distance, highway adaptability and price; S6. Dynamic weight adjustment step: Based on the real-time rainfall, dynamically adjust the weight of each recommendation factor and perform normalization processing. S7. Comprehensive scoring step: Based on the normalized weights and each recommendation factor, calculate the comprehensive recommendation degree for each of the effective candidate parking lots; S8. Output Step: Based on the comprehensive recommendation ranking, output the recommendation results and determine whether the safety recommendation factor of the recommended parking lot is lower than the preset safety threshold. If so, add a safety warning sign to the parking lot during output.

[0005] Specifically, the path planning steps include: Based on road network data, calculate the cost C for each road segment. i The calculation formula is as follows:

[0006] Among them, L i Let V be the length of road segment i. i Let R be the speed limit for road segment i, R be the real-time rainfall, and ω be the speed limit for road segment i. risk (R) represents the dynamic risk weight, and P... i (R) represents the probability of flooding in road segment i.

[0007] Specifically, the flooding probability P i The formula for calculating (R) is:

[0008] The formula for calculating the dynamic risk weight is as follows:

[0009] Where, β i Let be the flood risk coefficient for road segment i.

[0010] Specifically, the path planning step S3 uses a heuristic search algorithm, and the formula for calculating its heuristic function h(n) is as follows:

[0011] Among them, dist euclidean (n, goal) represents the Euclidean distance from the current node n to the target parking lot, V max The maximum speed limit for the road network is given by α, which is a weighting coefficient. This is the estimated average road risk coefficient in the straight line direction from node n to the target parking lot.

[0012] Specifically, the estimated average road risk coefficient Determined in the following ways: Determine the connection between the current node n and the target node goal; Create a buffer of a specified width along the connection line; Obtain all road segments that intersect with the buffer zone; Calculate the risk coefficient β for all the intersecting road segments. i The average value, as .

[0013] Specifically, the formula for calculating the weighted average flood probability is as follows:

[0014] in, For the weighted average probability of flooding, dpath P is the total path length. k (R) represents the probability of flooding in road segment k, and L k Let k be the length of road segment k.

[0015] Specifically, in the recommendation factor calculation step, each recommendation factor is calculated according to the following formula: The security recommendation factor:

[0016] The distance recommendation factor:

[0017] The high-speed adaptability recommendation factor:

[0018] The price recommendation factor:

[0019] Where p is the parking fee.

[0020] Specifically, in the weight dynamic adjustment step, the initial weight of each recommendation factor is adjusted according to the rainfall R according to the following rules: Safety weight ; Distance weight ); High-speed adaptability weight:

[0021] Price weighting:

[0022] And for ω s ω d ω h ω p Normalization is performed to obtain the final weight assignment ω' s ,ω' d ,ω' h ,ω' p .

[0023] Specifically, the formula for calculating the overall recommendation score RS of each valid candidate parking lot is as follows:

[0024] Furthermore, following step 8, the following is also included: S9: Real-time update step: Monitor the changes in rainfall ΔR and vehicle displacement Δd in real time. If ΔR ≥ 5 mm / h or Δd ≥ 1 km, return to the path planning step to recalculate.

[0025] The beneficial effects of this invention are as follows: By integrating real-time rainfall, road network risk coefficients, vehicle status, and parking lot information from multiple dimensions, this invention constructs a dynamic and adaptive path cost model and decision weight allocation mechanism. It can dynamically optimize path planning based on rainfall intensity, prioritize avoiding high-flood-risk road sections, and automatically adjust the weights of decision factors such as safety, distance, scenario adaptability, and economy according to the severity of weather conditions. This ensures that the recommended strategy prioritizes safety in extreme situations. At the same time, by distinguishing between highway and non-highway scenarios and accurately matching parking lot types, it significantly improves the rationality and practicality of the recommended results. Finally, through integrated ranking, safety warnings, and multimodal human-computer interaction guidance, it provides intelligent decision support for drivers in emergency situations, effectively improving driving safety and risk avoidance efficiency. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating the parking spot recommendation method of the present invention. Detailed Implementation

[0027] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The drawings are for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention.

[0028] In the process described in the specification, claims, or drawings of this invention, each step is numbered (e.g., step 10, 20, etc.). These numbers are used only to distinguish the steps and do not represent any execution order. It should be noted that the terms "first," "second," etc., used herein are only for distinguishing the objects being described and do not represent a chronological order, nor do they indicate that "first," "second," etc., are different types.

[0029] like Figure 1 As shown, an embodiment of the present invention provides a parking spot recommendation method, including: S101, Data Acquisition and Initialization.

[0030] The vehicle terminal obtains its status via the CAN bus, including its current latitude and longitude coordinates (x, y, y). v ,y v And determine the current road type (such as highway, urban expressway, ordinary road) through GPS and map matching.

[0031] Real-time rainfall (R, unit: mm / h) is obtained from a meteorological service terminal via an in-vehicle internet module or through a rain sensor installed on the vehicle.

[0032] Simultaneously, a request is sent to the cloud service to obtain information on all candidate parking lots within a preset search radius (e.g., 50km, configurable by the user) centered on the vehicle, including their coordinates (x, y, y). p ,y p ), ParkType (e.g., service area, emergency exit, near highway exit, normal), real-time available seats, price p, and distance d to the nearest highway exit. exit (Unit: km); and obtain road network data within this area from digital map service providers, including the length L of each road segment. i Speed ​​limit V i Flooding risk coefficient β i wait.

[0033] S102, Candidate parking lot filtering.

[0034] The cloud server or vehicle-side system performs initial filtering on the acquired list of candidate parking lots. First, parking lots with zero available spaces are removed. Second, the straight-line distance between each parking lot and the vehicle's current location is calculated, and parking lots with a distance greater than a user-defined threshold (e.g., 10km) are removed, forming a set of valid candidate parking lots.

[0035] S103, Optimal Path Planning.

[0036] For each valid candidate parking lot, the system executes a path planning algorithm to find an optimal path. The specific steps include: First, based on the real-time rainfall R and the flooding risk coefficient β for each road segment... i Calculate the probability of flooding on this road section:

[0037] Among them, the flood risk coefficient β i Provided by the server, it can be obtained based on the road's historical flooding history.

[0038] Then, the dynamic risk weights are calculated:

[0039] Next, calculate the comprehensive cost C for each road segment. i :

[0040] The cost takes into account both travel time and the risk of flooding.

[0041] This embodiment uses a hair-pattern search algorithm (such as the A* algorithm) for path planning, and the heuristic function h(n) is designed as follows:

[0042] Among them, dist euclidean (n,goal) represents the straight-line distance (Euclidean distance) from the current node n to the target parking lot location, V max Let α be the maximum permissible speed in the entire road network, and let α be a weighting coefficient, where α ∈ [0, 1]. This represents the estimated average road risk coefficient along a straight line from the current node n to the target point.

[0043] In this embodiment, the estimated average road risk coefficient Determined in the following ways: Determine the connection between the current node n and the target node goal; Create a buffer of a specified width along the connection line; Obtain all road segments that intersect with the buffer zone; Calculate the risk coefficient β for all the intersecting road segments. i The average value, as .

[0044] The algorithm ultimately outputs an optimal path consisting of a sequence of road segments.

[0045] S104, Path safety assessment.

[0046] A safety assessment is performed on the planned optimal path, including: First, extract the sequence of road segments from the optimal path, and traverse each road segment L. k Get the total path length d path :

[0047] Next, calculate the weighted average probability of flooding along the entire route:

[0048] The smaller the value, the higher the overall security of the path.

[0049] S105. Calculate the recommendation factor.

[0050] In this embodiment, four recommendation factors are calculated for each candidate parking lot: Safety recommendation factor:

[0051] Distance recommendation factor:

[0052] High-speed adaptability recommendation factor:

[0053] Price recommendation factor:

[0054] Where p is the parking fee.

[0055] S106. Calculate and normalize the weights of each recommendation factor dynamically based on the real-time rainfall R.

[0056] In this embodiment, the initial weights of each recommendation factor are adjusted according to the rainfall R as follows: Safety weight ; Distance weight ); High-speed adaptability weight ω h :

[0057] Price weight ω p :

[0058] Then, for ω s ω d ω h ω p Normalization is performed to obtain the final weight assignment ω' s ,ω' d ,ω' h ,ω' p The specific calculation method is as follows:

[0059] S107. Calculate the overall recommendation score.

[0060] The recommendation factors are weighted and summed using normalized weights to obtain the final overall recommendation score RS for each parking lot:

[0061] S108. Output the recommendation results.

[0062] Sort all valid candidate parking lots by RS value from highest to lowest, and select the top three parking lots as recommendations, outputting them to the vehicle's HMI interface. Simultaneously, check the RS value of each recommended parking lot. score If the value is below the safety threshold (e.g., 0.2), a warning icon (e.g., an exclamation mark) will be displayed next to the parking lot sign to indicate that the destination or route still has a high risk; otherwise, a recommended icon will be displayed.

[0063] In another embodiment of the invention, the method further includes the following step after step 108: S109, Loop Trigger Judgment.

[0064] The system continuously monitors the changes in rainfall ΔR and vehicle displacement Δd. If ΔR ≥ 5 mm / h or Δd ≥ 1 km is detected, it determines that the current recommended result may no longer be optimal and immediately triggers the process to recalculate from S103 to ensure the real-time performance and accuracy of the recommendation.

[0065] In practice, the monitoring frequency of rainfall changes can be set according to actual conditions, for example, to acquire instantaneous rainfall every 5-10 minutes. When judging rainfall changes, the latest instantaneous rainfall intensity R can be obtained from the cache or interface. c And the previous instantaneous rainfall R p Thus, the change in rainfall ΔR can be calculated: ΔR=|R c - R p | The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for recommending temporary parking spots, characterized in that, Includes the following steps: S1. Data collection steps: Obtain vehicle status information, real-time rainfall, candidate parking lot information within the search radius, and road network data; S2. Parking lot filtering step: Based on the candidate parking lot information, filter out parking lots with no vacancies or those whose distance exceeds a preset threshold to obtain a set of effective candidate parking lots. S3. Path planning step: For each of the valid candidate parking lots, calculate the optimal path from the vehicle's current location to the parking lot, and output the path segment sequence; S4. Safety assessment step: Based on the road segment sequence of the optimal route, calculate the weighted average flooding probability of the route; S5. Recommendation factor calculation steps: Calculate the recommendation factor for each of the effective candidate parking lots, wherein the recommendation factor includes at least safety, distance, highway adaptability and price; S6. Dynamic weight adjustment step: Based on the real-time rainfall, dynamically adjust the weight of each recommendation factor and perform normalization processing. S7. Comprehensive scoring step: Based on the normalized weights and each recommendation factor, calculate the comprehensive recommendation degree for each of the effective candidate parking lots; S8. Output Step: Based on the comprehensive recommendation ranking, output the recommendation results and determine whether the safety recommendation factor of the recommended parking lot is lower than the preset safety threshold. If so, add a safety warning sign to the parking lot during output.

2. The temporary parking spot recommendation method according to claim 1, characterized in that, The path planning steps include: Based on road network data, calculate the cost C for each road segment. i The calculation formula is as follows: Among them, L i Let V be the length of road segment i. i Let R be the speed limit for road segment i, R be the real-time rainfall, and ω be the speed limit for road segment i. risk (R) represents the dynamic risk weight, and P... i (R) represents the probability of flooding in road segment i.

3. The method according to claim 2, characterized in that, The probability of flooding P i The formula for calculating (R) is: The formula for calculating the dynamic risk weight is as follows: Where, β i Let be the flood risk coefficient for road segment i.

4. The temporary parking spot recommendation method according to claim 2 or 3, characterized in that, The path planning step S3 uses a heuristic search algorithm, and the formula for calculating its heuristic function h(n) is as follows: Among them, dist euclidean (n, goal) represents the Euclidean distance from the current node n to the target parking lot, V max The maximum speed limit for the road network is given by α, which is a weighting coefficient. This is the estimated average road risk coefficient in the straight line direction from node n to the target parking lot.

5. The temporary parking spot recommendation method according to claim 4, characterized in that, The estimated average road risk coefficient Determined in the following ways: Determine the connection between the current node n and the target node goal; Create a buffer of a specified width along the connection line; Obtain all road segments that intersect with the buffer zone; Calculate the risk coefficient β for all the intersecting road segments. i The average value, as .

6. The temporary parking spot recommendation method according to claim 3, characterized in that, The formula for calculating the weighted average probability of flooding is as follows: in, For the weighted average probability of flooding, d path P is the total path length. k (R) represents the probability of flooding in road segment k, and L k Let k be the length of road segment k.

7. The temporary parking spot recommendation method according to claim 6, characterized in that, In the recommendation factor calculation step, each recommendation factor is calculated according to the following formula: The security recommendation factor: The distance recommendation factor: The high-speed adaptability recommendation factor: The price recommendation factor: Where p is the parking fee.

8. The temporary parking spot recommendation method according to claim 1, characterized in that, In the aforementioned dynamic weight adjustment step, the initial weights of each recommendation factor are adjusted according to the rainfall R based on the following rules: Safety weight ; Distance weight ); High-speed adaptability weight Price weight And for ω s ω d ω h ω p Normalization is performed to obtain the final weight assignment ω' s ,ω' d ,ω' h ,ω' p .

9. The temporary parking spot recommendation method according to claim 8, characterized in that, The formula for calculating the overall recommendation score RS of each valid candidate parking lot is as follows: 。 10. The temporary parking spot recommendation method according to claim 1, characterized in that, The process after step 8 also includes: S9: Real-time update step: Monitor the changes in rainfall ΔR and vehicle displacement Δd in real time. If ΔR ≥ 5 mm / h or Δd ≥ 1 km, return to the path planning step to recalculate.

Images