CN114261399A - Decision planning method for intelligent driving of automobile under ice and snow road surface - Google Patents
Decision planning method for intelligent driving of automobile under ice and snow road surface Download PDFInfo
- Publication number
- CN114261399A CN114261399A CN202111403108.1A CN202111403108A CN114261399A CN 114261399 A CN114261399 A CN 114261399A CN 202111403108 A CN202111403108 A CN 202111403108A CN 114261399 A CN114261399 A CN 114261399A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- state
- planning
- lane
- ice
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 239000000203 mixture Substances 0.000 claims abstract description 12
- 230000008859 change Effects 0.000 claims description 16
- 238000002360 preparation method Methods 0.000 claims description 15
- 238000002156 mixing Methods 0.000 claims description 14
- 238000005070 sampling Methods 0.000 claims description 7
- 238000009825 accumulation Methods 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 5
- 230000009471 action Effects 0.000 claims description 3
- 230000001960 triggered effect Effects 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims 1
- 230000006399 behavior Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 230000001133 acceleration Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000006073 displacement reaction Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to a decision planning method for intelligent driving of an automobile under a snow and ice road surface, which mainly comprises the following steps: the method comprises the following steps: a decision-making method of a layered state machine is designed in an accumulated mode according to the lane changing safety distance and the dissatisfaction degree of drivers in the ice and snow environment; step two: improved blend A based on ice and snow environment*Planning a desired path by a planning rule; step three: and optimizing the generated path by adopting a cubic spline interpolation method, and planning a smooth track.
Description
Technical Field
The invention relates to design of a humanized advanced assisted driving system (ADAS), in particular to an intelligent driving automobile decision planning method based on environment.
Background
The ice and snow environment exists in the northeast and northwest regions of China and other vast regions for a long time, and vehicles in cold regions exist in the ice and snow environment for a long time, which is a special working condition. In northeast and northwest areas, the driver is long in winter, has low air temperature and large snowfall amount, and has poor driving conditions before removing snow. Some residual snow may also be left after the road is cleared of snow. Under the effect of ice and snow environment, the traffic running environment is severe, the road surface adhesion coefficient is only 1/8-1/4 of normal road surface, the driving comfort of a driver is poor, and great mental stress, fatigue and other discomfort are easy to generate. Under the ice and snow environment, the traffic operation efficiency and the driving safety are lower than those of a non-ice and snow road surface, and through analysis of highway traffic data in Harrison city, the average driving speed of vehicles under the ice and snow environment is lower than that of the non-ice and snow road surface by 20 percent. The factors cause many potential safety hazards, and data analysis of traffic accidents in 1967-.
The decision planning module in intelligent driving is one of the core indexes for measuring and evaluating the intellectualization of the intelligent driving automobile, and has the main tasks of analyzing the current environment after receiving various sensing information of the sensor and then giving an instruction to the bottom layer control module. The rule-based decision method is a classical method and was first widely used in the past DAPPA unmanned challenge race in 2004 and beyond. A Talos decision-making system of the Ma province's engineering university adopts a series state machine structure, the automatic driving system comprises positioning and navigation, obstacle detection, lane line detection, road sign identification, module planning modules of travelable area map construction, motion planning, motion control and the like, and a feasible path is generated by adopting a fast exploration random tree algorithm. In order to overcome the defects, the Stanford Junior unmanned vehicle firstly proposes to adopt a parallel structure state machine, each state is processed in a module mode, and the whole system can quickly and flexibly respond to input. However, under complex conditions, the algorithm structure becomes huge due to the increase of the traversal states, so that the division between the states and the solution of state conflicts become difficult. And the parallel structure is suitable for more complex working conditions. Compared with a series structure, the parallel structure has the advantages of wider scene traversal range, easier realization of complex function combination, better modularization and expandability and the like. The disadvantage is the lack of depth of the scene and slight variations in the environment can lead to decision errors. The learner combines the advantages of the serial connection type and the parallel connection type structures, and forever provides a layered decision-making method for decoupling the compliance with the traffic regulations and the avoidance of traffic participants, the adopted actions are selected according to the traffic regulations to form a candidate behavior set, then factors such as collision risks, traffic rules and the like are synthesized, an optimal behavior is comprehensively decided by adopting evaluation methods such as hidden Markov models to guess the intention of drivers of other vehicles, the driving efficiency, a macroscopic recommended path and the like, then path planning is carried out according to the behavior, the current region and the target region, and the recommended vehicle speed is given.
The intelligent driving automobile planning method is mainly divided into sampling-based planning and search-based planning. The sampling-based method has the problems of high solving complexity and low sampling efficiency in some boundary scenes. Zhang Yu et al applied a hybrid a-algorithm and a conjugate gradient descent path smoother for generating a smooth path by dynamically expanding four node patterns, introducing a kinematic model with respect to a-to make each edge mapped, i.e., the generated trajectory smoother, with the disadvantage of lacking an obstacle avoidance module and not considering the surrounding obstacles. Joge Godoy et al designs a local path planning module by using a path-based optimal decision method, the local path planning module interacts with a global path planning module, and has the functions of generating a smooth fitting track, avoiding static and dynamic obstacles and the like, and the problem of overhigh time complexity exists.
Disclosure of Invention
In the prior intelligent driving automobile decision planning research, a method aiming at a special working condition of ice and snow mixed road surface rarely appears, and the invention aims to provide a decision planning method aiming at an ice and snow environment.
In order to achieve the purpose, the invention adopts the following technical scheme:
a decision planning method for intelligent driving of automobiles under ice and snow road surfaces mainly relates to the following steps:
the method comprises the following steps: a decision framework of a layered state machine is designed in an accumulated mode according to the lane changing safety distance and the dissatisfaction degree of drivers in the ice and snow environment;
step two: improved blend A based on ice and snow environment*The planning algorithm plans out the expected path,
step three: optimizing the planned expected path by adopting a cubic spline interpolation method so as to obtain a smooth track;
2. designing a decision framework of a hierarchical state machine comprises the following steps:
1) a state machine decision model is designed, the model is divided into two layers, and the upper layer is divided into two states, namely normal driving and intersection. The lower layer is divided into three states, namely lane keeping, lane preparation, lane changing and stopping/waiting. The states of the two-layer state machine are stored in a key value pair, which indicates the current state of the vehicle, the formula (1) gives the state set of the state machine,
(Normal Driving, intersection, Lane keeping, preparation for Lane Change, stop/wait } (1)
The state transition relationship of the decision model is specifically as follows:
and (3) upper layer: the normal driving state is converted into an intersection state when the vehicle enters the intersection, and the intersection state is converted into a normal driving state after the vehicle leaves the intersection;
the lower layer: the stopping state is converted into a lane keeping state after being started, the lane changing state is converted into a lane changing preparation state when the dissatisfaction degree reaches a threshold value, the lane changing preparation state is judged according to whether lane changing is needed currently or not, whether a safety distance is met or not and whether collision occurs or not, and the lane changing preparation state is converted into a lane changing state (including left lane changing and right lane changing) when the conditions are met; after the vehicle reaches the target lane or the trajectory planning fails, the vehicle cannot reach the target lane, the lane changing state is converted into a lane keeping state, and the vehicle enters a stopping/waiting state when the vehicle reaches the ending condition and braking is performed under the lane keeping state;
2) considering that the running continuity and speed of the vehicle cannot meet the driver's expectation due to the poor adhesion condition of the ice and snow road surface and the presence of factors such as a low-speed preceding vehicle under the ice and snow environment, the dissatisfaction of the driver is gradually accumulated to generate the overtaking intention, and the driver dissatisfaction accumulation formula (2) is obtained.
Wherein D (k) is the dissatisfaction at this time; d (k-1) is the dissatisfaction degree of the last time point; v. ofpreA desired vehicle speed; v. ofdesTheta is the current running speed, theta is the running continuity weighting coefficient, CountavgT is the sampling time, which is the average number of stops per kilometer.
3) Considering the safety of lane changing or overtaking, whether the lane changing or overtaking can be carried out depends on the front and back positions of the traffic participants on the adjacent lanes, namely the lane changing safety distance. Considering the safe braking distance in the ice and snow environment, the safe braking distance can be divided into a braking action distance and a braking driving distance, and a front vehicle spacing distance, as shown in formula (3).
Wherein s is the safety braking distance s0Is the difference between the front and rear distances after braking, abmaxFor maximum value of braking deceleration, s0Is the difference between the front and rear vehicle distances, t1,t2V is the braking time and the speed of the vehicle.
3. Improved blend A based on ice and snow environment*The planning algorithm comprises the following steps:
1) under the working condition of mixing the ice and snow road surface and the normal road surface, A is mixed*Algorithm 2D search extends to 4D space [ x, y, theta, D ]]Where θ is the heading of the host vehicle and D represents whether the vehicle is moving forward or backward; inputting position information of an ice-snow road surface and vehicle track prediction information to a planning model, and if a searched node is covered by ice and snow, reducing the cost of the node; if mixing A*And if the track between the searched node and the current node is overlapped with the predicted track of the week vehicle, the accumulated cost on the track is reduced.
2) As shown in fig. 3, the distance between the self vehicle and the other vehicle is larger than the safety distance between the self vehicle and the front vehicle, and the speed expectation of the self vehicle and the other vehicle is met, the lane change condition is triggered, and the self vehicle passes throughMixing A*The algorithm searches out two tracks a, b. Since the track b node passes through the icy and snowy road surface, the overall cost is reduced, and therefore the planning module outputs the a track to the control module. As shown in fig. 4, the own vehicle triggers a lane change condition to search two tracks, a and b, and since the tracks a and b overlap with the tracks predicted by other vehicles on the right side of the own vehicle in a certain time period, that is, collision occurs, neither track a nor track b can be selected, a state of failed planning will be returned to the state machine module, and the state machine is shifted to a lane keeping state.
4. The method for optimizing the generated path by adopting the cubic spline interpolation method comprises the following steps:
1) the path is divided into several parts, each small part is interpolated to obtain different functions, and the interpolated functions are used to represent the whole curve.
2) Constructing cubic equations, i.e. S, for each parti(x)＝ai+bi(x-xi)+ci(x-xi)2+di(x-xi)3。
Wherein S isi(x) For piecewise interpolation function (i ═ 0,1, …, n), ai,bi,ci,diAs unknowns over the interpolation interval, xiAre interpolation points.
3) Let equation Si(x) Four interpolation conditions are satisfied. Condition one, the function must pass through all known nodes. Conditional two, the 0 th derivative must be continuous at the node. Condition three, 1 st order continuation at all nodes except the first and last nodes. Condition four, except for 2 order continuation at the first node and the last node.
Due to the adoption of the technical scheme, compared with the prior method, the method has the following advantages:
1. the invention designs a hierarchical state machine framework, which decouples the upper layer state and the lower layer state, wherein the upper layer is a scene, the lower layer is a sub-state in the scene, the upper layer state machine is used as a base class, and the lower layer state machine is used as a derivative class.
2. In the layered state machine framework, constraints such as an unsatisfied accumulated lane change mechanism and a driving safety distance in an ice and snow environment are introduced, and the safety and the comfort of a decision planning model in the ice and snow environment are improved.
3. The invention improves the mixing A based on the ice and snow environment*The method comprises the steps of planning an expected path by an algorithm, optimizing a track by a cubic spline interpolation method, introducing information of an ice and snow road surface and prediction information of a vehicle-surrounding track by the algorithm, and considering vehicle-surrounding behaviors and special working conditions under the ice and snow mixed road surface, thereby planning a safe and relatively smooth track.
Drawings
FIG. 1 is a schematic diagram of a hierarchical state machine decision model
FIG. 2 is a schematic view of a mixed ice and snow road surface
FIG. 3 is a trajectory planning scenario 1
FIG. 4 is a trajectory planning scenario 2
FIG. 5 is a schematic diagram of trajectory planning
FIG. 6 is a set simulation environment
FIG. 7 is a diagram of simulator setup parameters
FIG. 8 is a longitudinal displacement of a preceding vehicle and a following vehicle
FIG. 9 is a longitudinal displacement difference between a preceding vehicle and a following vehicle
FIG. 10 is a graph of the speed of the bicycle under a following condition
FIG. 11 is a graph showing the acceleration curve of the bicycle under the following driving condition
FIG. 12 is values of lane change status under overtaking condition
FIG. 13 is a curve of speed of the bicycle under overtaking condition
FIG. 14 is a curve of acceleration of a vehicle under overtaking condition
Detailed Description
The following further describes the details of the present invention and its embodiments.
The invention relates to a decision planning method for an intelligent driving automobile under an ice and snow road surface, which mainly comprises the following steps:
the method comprises the following steps: a decision-making method of a layered state machine is designed in an accumulated mode according to the lane changing safety distance and the dissatisfaction degree of drivers in the ice and snow environment; step two: improved blend A based on ice and snow environment*Generating a path by a planning algorithm; step three: and optimizing the generated path by adopting a cubic spline interpolation method, and planning a smooth track.
1 rule-based State machine decision model building
The decision model of the invention adopts a layered state machine, the upper layer is divided into two states, namely normal driving and intersection, and the lower layer is divided into four states, namely lane keeping, lane preparation, lane changing and stopping/waiting, and the state machine model is shown in figure 1. The transition between the states is determined by the safe distance between the self vehicle and the front vehicle in the ice and snow environment, the expected speed of the self vehicle on the ice and snow road surface and whether the planning module can plan a reasonable track.
{ Normal Driving, intersection, Lane keeping, preparation for Lane Change, stop/wait } (1)
The state transition relationship of the decision model is specifically as follows:
and (3) upper layer: the normal driving state is converted into an intersection state when the vehicle enters the intersection, and the intersection state is converted into a normal driving state after the vehicle leaves the intersection;
the lower layer: the stopping state is converted into a lane keeping state after being started, the lane changing state is converted into a lane changing preparation state when the dissatisfaction degree reaches a threshold value, the lane changing preparation state is judged according to whether lane changing is needed currently or not, whether a safety distance is met or not and whether collision occurs or not, and the lane changing preparation state is converted into a lane changing state (including left lane changing and right lane changing) when the conditions are met; and after the vehicle reaches the target lane or the trajectory planning fails and the vehicle cannot reach the target lane, switching the lane changing state into a lane keeping state, and entering a stopping/waiting state when the vehicle reaches the ending condition and braking in the lane keeping state.
The layered state machine architecture in the invention decouples the upper and lower layer states. The upper layer is a scene, and the lower layer is a sub-state in the scene; the upper layer state machine is used as a base class, and the lower layer state machine is used as a derived class. The framework has the advantage of convenient expansion, when a new scene needs to be responded or a scene task needs to be processed more carefully, a new state can be conveniently added in the original framework, the influence on the original various states is small, and the number of state transition conditions needing to be set is avoided by adopting a hierarchical state machine. In the related layered state machine framework, constraints such as an unsatisfied accumulated lane change mechanism and a driving safety distance in an ice and snow environment are introduced, and the safety and the comfort of a decision planning model in the ice and snow environment are improved.
1.1 dissatisfaction accumulation
Considering that in an ice and snow environment, due to poor ice and snow road surface adhesion conditions, a low-speed front vehicle and other factors exist, the running continuity and speed of the vehicle cannot meet the requirements of a driver, and when the vehicle runs on the ice and snow road surface, the driver does not have any intention of lane change due to sudden acceleration or deceleration. The vehicle speed is not too high due to poor ice and snow road adhesion conditions, and therefore the vehicle will have a high willingness to change lanes on ice and snow roads. Most of the reasons are that the front vehicle runs at a low speed for a long time, so that the passengers of the rear vehicle cannot meet the speed. The degree of dissatisfaction with the accumulation of the vehicle speed is set in a task file on the upper layer or set by other means, and an expected value of the vehicle speed during the driving process is set. But during actual driving the vehicle will follow the preceding vehicle at a very slow speed. The driving safety is ensured, but the expected driving speed is not reached. The cumulative degree of vehicle speed dissatisfaction will increase over time. The dissatisfaction degree of the driver is gradually accumulated to generate the overtaking intention, and a driver dissatisfaction degree accumulation formula (2) is obtained.
Wherein D (k) is the dissatisfaction degree at the current moment; d (k-1) is the dissatisfaction degree at the previous moment; v. ofpreA desired vehicle speed; v. ofdesTheta is the current running speed, theta is the running continuity weighting coefficient, CountavgT is the sampling period, which is the average number of stops per kilometer.
1.2 safety distance between bicycle and front bicycle
The minimum safe distance of the vehicle to the front vehicle means that an emergency stop occurs due to a sudden occurrence of the vehicle in front when the vehicle travels on a road. The vehicle can be braked within an effective and safe distance without causing traffic accidents. The target vehicle should travel a distance in excess of the minimum safe distance from the vehicle in front of the lane to prevent danger. The minimum safety distance can be obtained from the stopping distance. According to the analysis of the braking performance of the automobile, the distance has a direct relation with the driving safety of the automobile. From the whole braking process, the brake device comprises two parts: the distance traveled for the brake application time and the distance traveled for the continuous braking.
Wherein, abmaxFor maximum value of braking deceleration, s0Is the difference between the front and rear vehicle distances, t1,t2V is the braking time and the speed of the vehicle.
According to the safe distance between the self vehicle and the front vehicle and the establishment of the vehicle speed expectation formula, the transfer method of the state machine model can be obtained.
2 trajectory planning model establishment
The idea of the planning model establishment is to first adopt an improved mixture A*The algorithm determines the future position of the vehicle and a curve is fitted by a trajectory planning model of a cubic spline interpolation method for the controller to solve.
2.1 blend A*Planning method
Mixing A*The algorithm is divided into two stages, the first stage is actually the conventional A*The algorithm is modified, in contrast to the mixing A*The heuristic search is carried out under a continuous coordinate system, and the generated track can be ensured to meet the non-integrity constraint of the vehicle, but the path is not necessarily the global optimal path in the running process of the algorithm, and the path is in the vicinity of the global optimal solution. Tradition A*With mixing A*Both algorithms are based on the grid world, A*Blending A is given a corresponding penalty to the center points of each mesh and the algorithm only visits these center points*Points that satisfy the 3D continuity state of the vehicle are first chosen in these grids and the loss is assigned to these points. Different from the graph search algorithm A*Mixing A*Considering a kinematic model, A*Algorithm 2D search extends to 4D space [ x, y, theta, D ]]Where θ is the heading of the host vehicle and D represents whether the vehicle is moving forward or backward. In the figure, the searching method consists of six searching directions, and the arc radius is the minimum turning radius of the vehicle as shown in figure 5.
2.2 trajectory planning for specific conditions
The invention aims at the mixed working condition of the ice and snow road surface and the normal road surface and establishes a track planning model as shown in figure 2. Conventional blend A*The method does not consider the tracks of the ice and snow road surface and the surrounding vehicles, and the invention is used for the classical mixture A*And improving an algorithm, and inputting position information of the ice and snow road surface and vehicle track prediction information to the planning model. If the searched node is covered by ice and snow, the cost of the node is reduced. If mixing A*And if the track between the searched node and the current node is overlapped with the predicted track of the week vehicle, the accumulated cost on the track is reduced.
As shown in FIG. 3, the distance between the self vehicle and the other vehicle is larger than the safety distance between the self vehicle and the front vehicle and meets the speed expectation of the self vehicle and the other vehicle, the lane change condition is triggered, and the self vehicle passes through the mixture A*The algorithm searches out two tracks, namely a track a and a track b. Since the track b node passes through the icy and snowy road surface, the overall cost is reduced, and therefore the planning module outputs the a track to the control module. As shown in fig. 4, the self-vehicle triggers a lane change condition to search two tracks, a and b, and since the tracks a and b overlap with the tracks predicted by other vehicles on the right side of the self-vehicle in a certain time period, both the tracks a and b cannot be selected, a planning failure state is returned to the state machine module, and the state machine keeps following.
2.3 cubic spline sampling method
Due to the mixing of A*PlanningThe path is not smooth, and part of the path [ x, y ]]And outputting by adopting a track planning model based on a cubic spline interpolation method. And planning the track of the intelligent driving automobile. And planning a lane changing track according to the initial state and the target state of the vehicle, so that the vehicle reaches the target position at the appointed time.
The method for optimizing the generated path by adopting the cubic spline interpolation method comprises the following steps:
1) the path is divided into several parts, each small part is interpolated to obtain different functions, and the interpolated functions are used to represent the whole curve.
2) A cubic equation is constructed for each part, as in equation (4).
Si(x)＝ai+bi(x-xi)+ci(x-xi)2+di(x-xi)3 (4)
Wherein S isi(x) For piecewise interpolation function (i ═ 0,1, …, n), ai,bi,ci,diAs unknowns over the interpolation interval, xiAre interpolation points.
3) Let equation Si(x) Four interpolation conditions are satisfied. Condition one, the function must pass through all known nodes. Conditional two, the 0 th derivative must be continuous at the node. Condition three, 1 st order continuation at all nodes except the first and last nodes. Condition four, except for 2 order continuation at the first node and the last node.
It is necessary to find 4n equations to solve 4n unknowns, first because the interpolation condition S (x) must be satisfied for all pointsi)＝yi(i-0, 1, …, n), each of all n-1 interior points satisfying S except for two end pointsi(xi+1)＝yi+1If the front and rear piecewise cubic equations have 2(n-1) equations, and the two endpoints respectively satisfy the first and last cubic equations, the total number of the equations is 2 n; secondly, the first derivatives of the n-1 interior points should be continuous, i.e. the same point at the end of the i-th interval and at the beginning of the i + 1-th interval, and their first derivatives should also be equal, there are n-1 equations, and further the second derivatives of the interior pointsThe derivatives are also continuous, i.e. there are now a total of 4n-2 equations, the other two being derived by the boundary conditions.
The boundary condition is a natural boundary condition, and the second derivative of the specified endpoint is 0, namely S ″ (x)0)＝S”(xn) 0. The interpolation point is substituted into the interpolation function to obtain ai＝yiBy using hi＝xi+1-xiDenotes the step size, from Si(xi+1)＝yi+1Available formula (5)
From S'i(xi+1)＝S′i+1(xi+1) The formula (6) can be obtained.
S′i+1(xi+1)＝bi+1+2ci(xi+1-xi+1)+3di(xi+1-xi+1)2＝bi+1 (6)
From S ″)i(xi+1)＝S″i+1(xi+1) Can be pushed out 2ci+6hidi＝2ci+1. Let mi＝S”i(xi)＝2ciThen, the formula (7) is obtained.
A is toi,ci,diThe substitution into the formula (6) gives the formula (8).
At natural boundary conditions, m0＝0，m n0. From equation (8), a linear equation system with m as an unknown number can be obtained as equation (9).
The coefficient matrix on the left side is a strict diagonal dominance matrix, so that the linear equation set has a unique solution, and the solution of the equation set is substituted into the formula (4) to obtain:
the invention improves the mixing A based on the ice and snow environment*The method comprises the steps of planning an expected path by an algorithm, optimizing a track by a cubic spline interpolation method, introducing information of an ice and snow road surface and prediction information of a vehicle-surrounding track by the algorithm, and considering vehicle-surrounding behaviors and special working conditions under the ice and snow mixed road surface, thereby planning a safe and relatively smooth track.
3 Observation data acquisition and screening
3.1 simulation experiment design for simulating driving
The simulation driving device is designed by taking a simulation driver built on the basis of CarMaker vehicle dynamics simulation software as an experiment platform, a decision planning algorithm is verified by adopting various working conditions, and driving behaviors, vehicle states and road environment information are collected. The main software and hardware of the driving simulation platform are as follows:
the simulated driver is in the real form, and the Luo skill G27 suite comprises an accelerator pedal, a brake pedal, a clutch pedal, a gear handle, a steering wheel and the like which have the same driving position structure as the real vehicle. The acrobatic G27 steering wheel provides force feedback technology to provide road feel information to the driver, and the magnitude of the feedback force proportional coefficient can also be set. The scene can be customized through an IPGRoad module in a CarMaker main interface, and the invention establishes a straight-going-rotary-straight-going working condition through the IPGRoad. Vehicle parameters are set in the CIT of CarMaker software, and as shown in fig. 7, vehicle-related parameters and the progress of the driving strategy simulation can be controlled through the interface, the vehicle is selected as audio 8, the parameters such as tires are selected as default values, and the experimental setting is only driven by a driver, so that the load is set to 60kg here. The experimental lane is an urban road with three lanes, an ice and snow road surface is randomly inserted into the lane, and the friction coefficient of the road surface is 0.28. The road condition is shown in fig. 6, and the light gray road surface is the ice and snow covered road surface.
3.1.1 following Condition
The method comprises the steps that a front vehicle runs at 30km/h and is set to cruise at a constant speed, the front vehicle is located 15m in front of the self vehicle, the self vehicle starts from a static state and follows the front vehicle, the expected speed is set to be 30km/h, the distance between the self vehicle and the front vehicle is kept, the self vehicle accelerates from the static state to the running speed of 30km/h, and the speed of the self vehicle is controlled to be stable.
The decision of the self-vehicle is determined by a state machine, and the self-vehicle keeps cruising at a constant speed in the following process. Fig. 8 shows the longitudinal displacement of the bicycle and the front bicycle. The leading vehicle accelerates from a standstill to travel at 30km/h, the speed curve of the own vehicle being shown in fig. 10.
The following distance required to meet the safety requirement can be calculated by the formula (3) to be more than 20 m. As can be seen from fig. 9, the traveling distance between the front vehicle and the rear vehicle is reduced and then increased, and finally the heel safety distance is filled. The acceleration is as shown in figure 11, the minimum is-3.9, the maximum is 1.0, and the comfort requirement and the safety requirement are met.
3.1.2 Overtaking operating mode
Setting two vehicles in front to cruise at a constant speed of 20km/h with a distance of 50m, and setting a distance between a self vehicle and a front vehicle with a distance of 50m, determining whether to select lane change by a self vehicle decision planning module, generating a lane change intention when the dissatisfaction accumulation reaches a set value, and performing lane change decision and path planning by the self vehicle. During driving, the dissatisfaction degree of the front vehicle gradually accumulates to reach a threshold value, the state machine is switched from a lane keeping state to a lane changing state as shown in figure 12, the speed of the self vehicle is shown in figure 13, and the acceleration of the self vehicle is shown in figure 14. The minimum is-4.3, and the maximum is 2.0, so that the comfort requirement and the safety requirement are met.
Claims (5)
1. A decision planning method for intelligent driving of automobiles under ice and snow road surfaces mainly relates to the following steps:
the method comprises the following steps: a decision framework of a layered state machine is designed in an accumulated mode according to the lane changing safety distance and the dissatisfaction degree of drivers in the ice and snow environment;
step two: improved blend A based on ice and snow environment*The planning algorithm plans out the expected path,
step three: optimizing the planned expected path by adopting a cubic spline interpolation method so as to obtain a smooth track;
the decision model of the state machine is divided into two layers, wherein the upper layer is divided into two states, namely normal driving and intersection, and the lower layer is divided into four states, namely lane keeping, lane preparation, lane changing and stopping/waiting; the states of the two-layer state machine are stored in a key value pair, which indicates the current state of the vehicle, the formula (1) gives the state set of the state machine,
(Normal Driving, intersection, Lane keeping, preparation for Lane Change, stop/wait } (1)
The state transition relationship of the decision model is specifically as follows:
and (3) upper layer: the normal driving state is converted into an intersection state when the vehicle enters the intersection, and the intersection state is converted into a normal driving state after the vehicle leaves the intersection;
the lower layer: the stopping state is converted into a lane keeping state after being started, when the dissatisfaction reaches a threshold value, the stopping state is converted into a lane changing preparation state, the lane changing preparation state is judged according to whether lane changing is needed currently or not, whether a safety distance is met or not and whether collision occurs or not, and when the condition is met, the lane changing preparation state is converted into the lane changing state; after the vehicle reaches the target lane or the trajectory planning fails, the vehicle cannot reach the target lane, the lane changing state is converted into a lane keeping state, and the vehicle enters a stopping state when the vehicle reaches the ending condition and is braked under the lane keeping state;
step two, the improved mixing A based on the ice and snow environment*The planning algorithm is to combine A with the normal road surface under the condition of mixing the ice and snow road surface*Algorithm 2D search extends to 4D space [ x, y, theta, D ]]Where θ is the heading of the host vehicle and D represents whether the vehicle is moving forward or backward; inputting position information of an ice-snow road surface and vehicle track prediction information to a planning model, and if a searched node is covered by ice and snow, reducing the cost of the node; if mixing A*And if the track between the searched node and the current node is overlapped with the predicted track of the week vehicle, the accumulated cost on the track is reduced.
2. The decision planning method for intelligently driving the automobile under the icy and snowy road surface according to claim 1, wherein the method for accumulating the dissatisfaction degree of the driver is as follows:
a driver dissatisfaction degree accumulation formula (2);
wherein D (k) is the dissatisfaction at this time; d (k-1) is the dissatisfaction degree of the last time point; v. ofpreA desired vehicle speed; v. ofdesTheta is the current running speed, theta is the running continuity weighting coefficient, CountavgT is the sampling time, which is the average number of stops per kilometer.
3. The decision planning method for intelligent driving under snowy and icy road surface as claimed in claim 1, wherein the lane-changing safety distance is divided into a braking action distance and a braking driving distance, and a front vehicle separation distance, as shown in formula (3),
wherein s is the safety braking distance s0Is the difference between the front and rear distances after braking, abmaxFor maximum value of braking deceleration, s0Is the difference between the front and rear vehicle distances, t1,t2V is the braking time and the speed of the vehicle.
4. The decision planning method for intelligently driving automobiles under icy and snowy roads according to claim 1, wherein the second step is to improve the mixture A based on the icy and snowy environment*The specific method for planning the expected path by the planning algorithm is as follows:
the distance between the self vehicle and the other vehicles is larger than the safety distance between the self vehicle and the front vehicle, the speed expectation of the self vehicle and the other vehicles is met, the lane change condition is triggered, and the self vehicle passes through the mixture A*Searching two tracks a and b by an algorithm; due to the trackThe node b passes through the ice and snow road surface, so the total cost is reduced, and the planning module outputs a track to the control module; the self vehicle triggers a lane changing condition to search out two tracks, a and b, and the two tracks, a and b, can be overlapped with the predicted tracks of other vehicles on the right side of the self vehicle in a certain time period, namely collision occurs, so that the two tracks, a and b, can not be selected, a planning failure state is returned to the state machine module, and the state machine is transferred to a lane keeping state.
5. The decision planning method for intelligently driving the automobile under the icy and snowy road surface according to claim 1 is characterized by comprising the following steps: the method for optimizing the planned expected path by adopting the cubic spline interpolation method to obtain the smooth track specifically comprises the following steps:
1) dividing the path into a plurality of parts, interpolating each small part to obtain different functions, and representing the overall curve by the interpolated functions;
2) constructing cubic equations, i.e. S, for each parti(x)＝ai+bi(x-xi)+ci(x-xi)2+di(x-xi)3；
Wherein S isi(x) For piecewise interpolation function (i ═ 0,1, …, n), ai,bi,ci,diAs unknowns over the interpolation interval, xiIs an interpolation point;
3) let equation Si(x) Four interpolation conditions are satisfied; condition one, the function must pass through all known nodes; condition two, the 0 th derivative must be continuous at the node; condition three, all nodes except the first node and the last node are 1-order continuous; condition four, except for 2 order continuation at the first node and the last node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111403108.1A CN114261399B (en) | 2021-11-24 | 2021-11-24 | Decision planning method for intelligent driving automobile under ice and snow road surface |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111403108.1A CN114261399B (en) | 2021-11-24 | 2021-11-24 | Decision planning method for intelligent driving automobile under ice and snow road surface |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114261399A true CN114261399A (en) | 2022-04-01 |
| CN114261399B CN114261399B (en) | 2023-11-07 |
Family
ID=80825479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111403108.1A Active CN114261399B (en) | 2021-11-24 | 2021-11-24 | Decision planning method for intelligent driving automobile under ice and snow road surface |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114261399B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116620287A (en) * | 2023-06-30 | 2023-08-22 | 南京项尚车联网技术有限公司 | Intelligent driving method and system |
| DE102023116536B3 (en) | 2023-06-23 | 2024-07-18 | Audi Aktiengesellschaft | Method for operating an autonomously controlled vehicle, autonomously controllable motor vehicle and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101297337A (en) * | 2005-09-29 | 2008-10-29 | 微软公司 | Methods for predicting destinations from partial trajectories employing open-and closed-world modeling methods |
| CN108454623A (en) * | 2018-01-22 | 2018-08-28 | 大连理工大学 | A kind of unmanned electric vehicle Trajectory Tracking Control method of four motorized wheels |
| US20190113919A1 (en) * | 2017-10-18 | 2019-04-18 | Luminar Technologies, Inc. | Controlling an autonomous vehicle using smart control architecture selection |
| CN110304074A (en) * | 2019-06-28 | 2019-10-08 | 电子科技大学 | A kind of hybrid type driving method based on stratification state machine |
| CN111890951A (en) * | 2020-08-07 | 2020-11-06 | 吉林大学 | Intelligent electric automobile trajectory tracking and motion control method |
| US20200406928A1 (en) * | 2019-06-28 | 2020-12-31 | Robert Bosch Gmbh | Method for controlling a vehicle |
| WO2021102955A1 (en) * | 2019-11-29 | 2021-06-03 | 华为技术有限公司 | Path planning method for vehicle and path planning apparatus for vehicle |
-
2021
- 2021-11-24 CN CN202111403108.1A patent/CN114261399B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101297337A (en) * | 2005-09-29 | 2008-10-29 | 微软公司 | Methods for predicting destinations from partial trajectories employing open-and closed-world modeling methods |
| US20190113919A1 (en) * | 2017-10-18 | 2019-04-18 | Luminar Technologies, Inc. | Controlling an autonomous vehicle using smart control architecture selection |
| CN108454623A (en) * | 2018-01-22 | 2018-08-28 | 大连理工大学 | A kind of unmanned electric vehicle Trajectory Tracking Control method of four motorized wheels |
| CN110304074A (en) * | 2019-06-28 | 2019-10-08 | 电子科技大学 | A kind of hybrid type driving method based on stratification state machine |
| US20200406928A1 (en) * | 2019-06-28 | 2020-12-31 | Robert Bosch Gmbh | Method for controlling a vehicle |
| WO2021102955A1 (en) * | 2019-11-29 | 2021-06-03 | 华为技术有限公司 | Path planning method for vehicle and path planning apparatus for vehicle |
| CN111890951A (en) * | 2020-08-07 | 2020-11-06 | 吉林大学 | Intelligent electric automobile trajectory tracking and motion control method |
Non-Patent Citations (1)
| Title |
|---|
| 熊璐;康宇宸;张培志;朱辰宇;余卓平;: "无人驾驶车辆行为决策系统研究", 汽车技术, no. 08 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102023116536B3 (en) | 2023-06-23 | 2024-07-18 | Audi Aktiengesellschaft | Method for operating an autonomously controlled vehicle, autonomously controllable motor vehicle and system |
| CN116620287A (en) * | 2023-06-30 | 2023-08-22 | 南京项尚车联网技术有限公司 | Intelligent driving method and system |
| CN116620287B (en) * | 2023-06-30 | 2024-03-19 | 南京项尚车联网技术有限公司 | Intelligent driving method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114261399B (en) | 2023-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6451111B2 (en) | Driving support device and driving support method | |
| CN111867911B (en) | Vehicle control method and apparatus | |
| WO2021136130A1 (en) | Trajectory planning method and apparatus | |
| JP3714258B2 (en) | Recommended operation amount generator for vehicles | |
| Mensing et al. | Trajectory optimization for eco-driving taking into account traffic constraints | |
| CN110304074B (en) | Hybrid driving method based on layered state machine | |
| CN103347757B (en) | Optimize the system and method for fuel economy by using prediction environment and driving behavior information | |
| Khodayari et al. | A historical review on lateral and longitudinal control of autonomous vehicle motions | |
| Bertolazzi et al. | Supporting drivers in keeping safe speed and safe distance: the SASPENCE subproject within the European framework programme 6 integrating project PReVENT | |
| CN114261399B (en) | Decision planning method for intelligent driving automobile under ice and snow road surface | |
| CN104925050B (en) | Traveling segmentation based on driving mode energy expenditure | |
| CN113297721B (en) | Simulation method and device for signal intersection vehicle selective exit road | |
| Kural et al. | Traffic light assistant system for optimized energy consumption in an electric vehicle | |
| CN108489500A (en) | A kind of global path planning method and system based on Energy Consumption Economy | |
| Aryal | Optimization of geometric road design for autonomous vehicle | |
| Boubaker et al. | Impact of intersection type and a vehicular fleet's hybridization level on energy consumption and emissions | |
| Chen et al. | Towards human-like speed control in autonomous vehicles: A mountainous freeway case | |
| CN117315970A (en) | Lane change of autonomous vehicles involving traffic congestion at an intersection | |
| Widaa et al. | Design of Fuzzy-based autonomous car control system | |
| Levitan et al. | Human factors design of automated highway systems | |
| Kamal et al. | Eco-driving using real-time optimization | |
| CN110426215B (en) | Model establishing method for vehicle ride comfort test and intelligent driving system | |
| CN117782126A (en) | Automatic driving path planning decision-making method guided by high-precision map | |
| Jiang | Research on the development of autonomous race cars and impact on self-driving cars | |
| Mihály et al. | Analysis of driver behavior related to look-ahead control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |