Path planning method for unmanned vehicle
By combining road slope and vehicle friction resistance, the shortest path function of the vehicle is constructed, the problem of insufficient accuracy of driverless cars is solved, the robustness and real-timeness of path planning is improved, and driving safety is improved.
Patent Information
- Application Number
- CN202411322361.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
Smart Images

Figure BDA0005054355590000021 
Figure BDA0005054355590000024 
Figure BDA0005054355590000031
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of unmanned vehicles, and more specifically, to a path planning method for an unmanned vehicle. Background Art
[0002] With the development of urbanization, the automobile industry has also achieved rapid development. However, with the annual increase in the number of cars, it has also caused a series of problems, such as increased traffic congestion and frequent traffic accidents, which in turn make the traffic environment increasingly harsh. Therefore, the development of driverless smart cars provides a new solution to solve the above problems.
[0003] Unmanned intelligent vehicles refer to vehicles that can drive autonomously without human intervention using artificial intelligence technologies such as computers, sensors, communications and control technologies. This technology is the implementation of a complex system based on technologies such as deep learning, machine vision and automatic control. Its goal is to provide a more efficient, more convenient and more environmentally friendly mode of transportation without sacrificing safety and stability.
[0004] However, most of the existing driverless smart cars are still only in the test stage, and most of the actual car applications require drivers to supervise and take over in a timely manner. Therefore, it is reflected that the existing driverless smart cars are unable to adapt to real environmental changes and traffic conditions. Summary of the invention
[0005] The purpose of the present invention is to design and develop a path planning method for an unmanned vehicle, which improves the accuracy and practicability of the path planning of the unmanned vehicle by combining the road slope and the vehicle friction resistance.
[0006] The technical solution provided by the present invention is:
[0007] A path planning method for an unmanned vehicle comprises the following steps:
[0008] Step 1: Collect the real-time position of the vehicle, the slope of the road ahead and the rolling resistance coefficient of the vehicle according to the sampling period;
[0009] Step 2: Determine the map based on the current real-time position of the vehicle, rasterize the map, use the real-time position of the vehicle as the starting position, and determine the end position and obstacle position;
[0010] Step 3: Create open and closed lists:
[0011] Put the obstacle locations and nodes that are inaccessible to vehicles into the closed list, and put the rest of the locations into the open list;
[0012] Step 4: Construct the shortest path function for vehicles:
[0013]
[0014] In the formula, F i is the normalized vehicle shortest path function, β is the first adjustment factor, γ is the second adjustment factor, is the cost from the starting position to the end position based on the grid, D0 is the Euclidean distance from the center of the starting position to the center of the end position, is the energy lost by the vehicle from the starting position to the end position based on the grid, E0 is the energy lost by the vehicle on the road with a distance of D0 and a rolling resistance coefficient of ψ′;
[0015] Step 5: With the starting position as the current node, traverse the surrounding nodes of the current node. If the surrounding nodes are in the open list, determine the node corresponding to the minimum value of the vehicle shortest path function among the surrounding nodes and update it as the current node. Delete the previous node from the open list and add it to the closed list.
[0016] Step 6: Traverse all nodes in the open list until the end point is reached to obtain the optimal path for the driverless vehicle.
[0017] Preferably, the sampling period is 10 seconds.
[0018] Preferably, the unit length of the grid is 1 m.
[0019] Preferably, the value ranges of the first adjustment factor and the second adjustment factor are both [0,1], and β+γ=1.
[0020] Preferably, the cost from the grid-based starting position to the end position satisfies:
[0021]
[0022] In the formula, is the cost from the grid-based starting position to the i-th grid position, is the cost from the grid-based i-th grid position to the end position.
[0023] Preferably, the cost from the grid-based starting position to the i-th grid position satisfies:
[0024]
[0025] In the formula, is the cost from the grid-based starting position to the i-1th grid position, ||i-1,i|| is the Euclidean distance from the i-1th grid position to the i-th grid position, θ i-1 is the road slope at the i-1th grid position, θ iis the road slope at the i-th grid position.
[0026] Preferably, the grid-based cost from the ith grid position to the end position satisfies:
[0027]
[0028] Where ||i,end|| is the Euclidean distance between the i-th grid position and the end point position based on the grid.
[0029] Preferably, the vehicle energy loss from the grid-based starting position to the end position satisfies:
[0030]
[0031] In the formula, is the vehicle energy loss from the grid-based starting position to the end position, is the vehicle energy loss from the grid-based starting position to the i-th grid position, is the energy loss of the vehicle from the i-th grid position to the end position.
[0032] Preferably, the vehicle loss energy from the i-th grid position to the end position satisfies:
[0033]
[0034] Where G is the weight of the vehicle and ψ′ is the average rolling resistance coefficient of the vehicle.
[0035] Preferably, the vehicle energy loss from the grid-based starting position to the i-th grid position satisfies:
[0036]
[0037] In the formula, ψ i-1 The rolling resistance coefficient of the vehicle at the i-1th grid position.
[0038] The beneficial effects of the present invention are:
[0039] A path planning method for an unmanned vehicle designed and developed by the present invention constructs a vehicle shortest path function by taking into account factors such as road slope and vehicle driving resistance, thereby improving the path planning accuracy and practicality of the unmanned vehicle, being more in line with the application of vehicle driving on actual roads, improving the robustness and real-time performance of path planning, and further improving the driving safety of the vehicle. DETAILED DESCRIPTION
[0040] The present invention is further described in detail below so that those skilled in the art can implement it according to the description.
[0041] The present invention provides a method for planning a path for an unmanned vehicle, comprising the following steps:
[0042] Step 1: Collect the real-time position of the vehicle, the slope of the road ahead and the rolling resistance coefficient of the vehicle according to the sampling period;
[0043] In this embodiment, the sampling period is 10 seconds;
[0044] Step 2: Determine the map based on the current real-time position of the vehicle, and rasterize the map, using the real-time position of the vehicle as the starting position, and determine the end position and obstacle position at the same time;
[0045] In this embodiment, the unit length of the grid is 1m;
[0046] Step 3: Create open and closed lists:
[0047] Put the obstacle positions and nodes that are inaccessible to vehicles into the closed list, and put the remaining positions into the open list, including the starting position;
[0048] Step 4: Construct the shortest path function for vehicles:
[0049] In actual conditions, searching for the shortest path for a vehicle is not necessarily the path with the least energy consumed by the vehicle. When the vehicle is driving normally, it will be subject to friction resistance, ramp resistance, and air resistance. For driverless vehicles, in order to simplify the amount of calculation in the algorithm, the air resistance of the driverless vehicle is much lower than the friction resistance and ramp resistance due to the limitation of driving speed. Therefore, the air resistance can be ignored. Therefore, driverless vehicles only need to consider ramp resistance and friction resistance.
[0050] First, considering only the slope resistance, mark the corresponding slope of each grid in the rasterized map according to the actual situation, thereby obtaining the shortest slope function from the starting position to the end position:
[0051]
[0052] In the formula, is the cost from the starting position to the end position based on the grid, m; is the cost from the grid-based starting position to the i-th grid position (current position), is the cost from the grid-based i-th grid position to the end position;
[0053] In a rasterized map, the cost refers to the grid distance;
[0054] Among them, the cost from the current position to the end position based on the grid satisfies:
[0055]
[0056] Where ||i,end|| is the Euclidean distance between the i-th grid position and the end point position based on the grid;
[0057] The cost from the grid-based starting position to the current position satisfies:
[0058]
[0059] In the formula, is the cost from the grid-based starting position to the i-1th grid position, and D(i-1,i) is the cost from the i-1th grid position to the i-th grid position;
[0060] In actual situations, the slope change of the road can be divided into two cases: the slope of the i-1th grid position is the same as or different from the slope of the i-th grid position. Specifically:
[0061] If the slope of the i-1th grid position is the same as that of the i-th grid position, then the cost from the i-1th grid position to the i-th grid position satisfies:
[0062]
[0063] Where ||i-1,i|| is the Euclidean distance from the i-1th grid position to the i-th grid position, and θ1 is the road slope;
[0064] If the slopes of the i-1th grid position and the i-th grid position are different, then the cost from the i-1th grid position to the i-th grid position satisfies:
[0065]
[0066] Where θ2 is the road slope at the i-1th grid position, and θ3 is the road slope at the i-th grid position;
[0067] In summary, the grid position and slope are combined to form i-1 Replace θ2 with θ i Replace θ3, then the cost from the i-1th grid position to the i-th grid position satisfies:
[0068]
[0069] Therefore, the cost from the starting position to the current position based on the grid becomes:
[0070]
[0071] Secondly, combining the ramp resistance and friction resistance, each grid in the rasterized map is marked with the corresponding rolling resistance coefficient according to the actual situation, and the vehicle loss energy function from the starting position to the end position is obtained:
[0072]
[0073] In the formula, is the vehicle loss energy from the grid-based starting position to the end position, J; is the vehicle energy loss from the grid-based starting position to the i-th grid position (current position), is the vehicle energy loss from the i-th grid position (current position) to the end position;
[0074] Among them, the vehicle loss energy from the i-th grid position to the end position satisfies:
[0075]
[0076] Where G is the weight of the vehicle, N; ψ′ is the average rolling resistance coefficient of the vehicle;
[0077] Among them, the average rolling resistance coefficient of the vehicle satisfies:
[0078]
[0079] In the formula, ψ i is the rolling resistance coefficient of the vehicle at the i-th grid position, ψ end is the vehicle rolling resistance coefficient at the end position;
[0080] The vehicle energy loss from the grid-based starting position to the i-th grid position satisfies:
[0081]
[0082] In the formula, is the vehicle energy loss from the grid-based starting position to the i-1th grid position, E(i-1,i) is the vehicle energy loss from the i-1th grid position to the i-th grid position;
[0083] Because of the slope factor, the value of E(i-1,i) includes the following cases:
[0084] 1. The i-1th grid position and the i-th grid position are both downhill sections, and the vehicle gravity is the power component. Therefore, the vehicle loss energy from the i-1th grid position to the i-th grid position satisfies:
[0085]
[0086] In the formula, ψ i-1 The rolling resistance coefficient of the vehicle at the i-1th grid position;
[0087] 2. The i-1th grid position is a downhill section, and the i-th grid position is an uphill section. The downhill section is the same as the first case above. The vehicle energy loss on the uphill section is affected by the slope resistance and road rolling resistance. The vehicle energy loss from the i-1th grid position to the i-th grid position satisfies:
[0088]
[0089] 3. The i-1th grid position and the i-th grid position are both uphill sections. The vehicle is always affected by the slope resistance and road rolling resistance. The vehicle loss energy from the i-1th grid position to the i-th grid position satisfies:
[0090]
[0091] 4. The i-1th grid position is an uphill section, and the i-th grid position is a downhill section. The vehicle is affected by the slope resistance and the road rolling resistance on the uphill section, but on the downhill section, the vehicle is only affected by the road rolling resistance. The vehicle loss energy from the i-1th grid position to the i-th grid position satisfies:
[0092]
[0093] In summary, the road slope is limited: uphill section θ i ≥0, downhill section θ i ≤0, then the vehicle energy loss from the i-1th grid position to the i-th grid position can be summarized as:
[0094]
[0095] At this time, the vehicle energy loss from the grid-based starting position to the end position satisfies:
[0096]
[0097] Thus, the normalized vehicle shortest path function is obtained to satisfy:
[0098]
[0099] In the formula, F i is the normalized vehicle shortest path function, β is the first adjustment factor, γ is the second adjustment factor, and the value range of the first adjustment factor and the second adjustment factor are both [0,1], and β+γ=1, D0 is the Euclidean distance from the center of the starting position to the center of the end position, and E0 is the energy loss of the vehicle on the road with a distance of D0 and a rolling resistance coefficient of ψ′;
[0100] Among them, the first adjustment factor and the second adjustment factor are different proportions of the shortest slope function from the starting position to the end position and the vehicle loss energy function from the starting position to the end position determined according to different situations, and when β=1, γ=0, the normalized vehicle shortest path function is the shortest slope function from the starting position to the end position, and when β=0, γ=1, the normalized vehicle shortest path function is the vehicle loss energy function from the starting position to the end position.
[0101] In this embodiment, the vehicle rolling resistance coefficient is different under different road conditions. On a straight and good road surface of concrete or asphalt, the vehicle rolling resistance coefficient is 0.01-0.05; on a gravel road with a good level, the vehicle rolling resistance coefficient is 0.05-0.1; on a dirt road with a poor level, the vehicle rolling resistance coefficient is 0.1-0.3; once the road changes more, the vehicle rolling resistance coefficient can be adjusted according to the actual situation.
[0102] Step 5: With the starting position as the current node, traverse the surrounding nodes of the current node to see if they are in the open list. If the surrounding nodes are in the open list, determine the node corresponding to the minimum value of the vehicle shortest path function among the surrounding nodes and update it as the current node, delete the previous node from the open list and add it to the closed list;
[0103] Step 6: Traverse all nodes in the open list until the end point is reached to obtain the optimal path for the driverless vehicle.
[0104] A path planning method for an unmanned vehicle designed and developed by the present invention constructs a vehicle shortest path function by taking into account factors such as road slope and vehicle driving resistance, thereby improving the path planning accuracy and practicality of the unmanned vehicle, being more in line with the application of vehicle driving on actual roads, improving the robustness and real-time performance of path planning, and further improving the driving safety of the vehicle.
[0105] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the embodiments shown and described herein.
Claims
1. A path planning method for an unmanned vehicle, characterized in that: The steps include: Step 1: Collect the real-time position of the vehicle, the slope of the road ahead and the rolling resistance coefficient of the vehicle according to the sampling period; Step 2: Determine the map based on the current real-time position of the vehicle, rasterize the map, use the real-time position of the vehicle as the starting position, and determine the end position and obstacle position; Step 3: Create open and closed lists: Put the obstacle locations and nodes that are inaccessible to vehicles into the closed list, and put the rest of the locations into the open list; Step 4: Construct the shortest path function for vehicles: In the formula, F i is the normalized vehicle shortest path function, β is the first adjustment factor, γ is the second adjustment factor, is the cost from the starting position to the end position based on the grid, D0 is the Euclidean distance from the center of the starting position to the center of the end position, is the energy lost by the vehicle from the starting position to the end position based on the grid, E0 is the energy lost by the vehicle on the road with a distance of D0 and a rolling resistance coefficient of ψ′; Step 5: With the starting position as the current node, traverse the surrounding nodes of the current node. If the surrounding nodes are in the open list, determine the node corresponding to the minimum value of the vehicle shortest path function among the surrounding nodes and update it as the current node. Delete the previous node from the open list and add it to the closed list. Step 6: Traverse all nodes in the open list until the end point is reached to obtain the optimal path for the driverless vehicle.
2. The path planning method for an unmanned vehicle according to claim 1, characterized in that: The sampling period is 10 seconds.
3. The path planning method for an unmanned vehicle according to claim 2, characterized in that: The unit length of the grid is 1 m.
4. The path planning method for an unmanned vehicle according to claim 3, characterized in that: The value ranges of the first adjustment factor and the second adjustment factor are both [0,1], and β+γ=1.
5. The path planning method for an unmanned vehicle according to claim 4, characterized in that: The cost from the grid-based starting position to the end position satisfies: In the formula, is the cost from the grid-based starting position to the i-th grid position, is the cost from the grid-based i-th grid position to the end position.
6. The path planning method for an unmanned vehicle according to claim 5, characterized in that: The cost from the grid-based starting position to the i-th grid position satisfies: In the formula, is the cost from the grid-based starting position to the i-1th grid position, ||i-1,i|| is the Euclidean distance from the i-1th grid position to the i-th grid position, θ i-1 is the road slope at the i-1th grid position, θ i is the road slope at the i-th grid position.
7. The path planning method for an unmanned vehicle according to claim 6, characterized in that: The cost from the grid-based i-th grid position to the end position satisfies: Where ||i,end|| is the Euclidean distance between the i-th grid position and the end point position based on the grid.
8. The path planning method for an unmanned vehicle according to claim 7, characterized in that: The vehicle energy loss from the grid-based starting position to the end position satisfies: In the formula, is the vehicle energy loss from the grid-based starting position to the end position, is the vehicle energy loss from the grid-based starting position to the i-th grid position, is the energy loss of the vehicle from the i-th grid position to the end position.
9. The path planning method for an unmanned vehicle according to claim 8, characterized in that: The vehicle energy loss from the i-th grid position to the end position satisfies: Where G is the weight of the vehicle and ψ′ is the average rolling resistance coefficient of the vehicle.
10. The path planning method for an unmanned vehicle according to claim 9, characterized in that: The vehicle energy loss from the grid-based starting position to the i-th grid position satisfies: In the formula, ψ i-1 The rolling resistance coefficient of the vehicle at the i-1th grid position.
