Disclosure of Invention
An object of the present invention is to solve at least the above problems and to provide at least the advantages described later.
The invention also aims to provide an intelligent decision-making method for avoiding parallel driving, which is applied to an automatic driving high-speed scene, and can optimize the upper limit of the speed required by the vehicle for staggered parallel driving after the vehicle and other vehicles are judged in advance, so that the vehicle speed of the vehicle can be adjusted in advance, and the situation that the automatic driving vehicle has long-time parallel driving at high speed is effectively avoided.
In order to achieve the above objects and other objects, the present invention adopts the following technical solutions:
an intelligent decision-making method for avoiding parallel driving applied to an automatic driving high-speed scene comprises the following steps:
the method comprises the steps of acquiring dynamic information of a vehicle, a driving road of the vehicle and vehicles around the vehicle in real time, and pre-judging whether other vehicles closest to the vehicle belong to a scene of parallel driving according to the collected dynamic information;
if so, optimizing the target speed upper limit of the vehicle through an evaluation function and a constraint condition, and adjusting the driving speed plan of the vehicle in real time according to the target speed upper limit;
if not, continuing to run according to the current running speed plan of the vehicle;
the scene of prejudging whether other vehicles closest to the vehicle belong to parallel driving according to the collected dynamic information comprises the following steps:
and calculating the time value of other vehicles, which are closest to the vehicle, on the two lanes from the dynamic information and leave the parallel driving area of the vehicle, and when the time value exceeds a preset time threshold value of staggering the parallel driving area, prejudging the situation that the vehicle and the other vehicles are driven in parallel, and setting the other vehicles as parallel driving vehicles.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the evaluation function includes:
calculating a time parameter according to the data of the other vehicle and a preset target speed upper limit;
calculating a speed difference value parameter for changing the upper limit of the speed according to the data of the vehicle, the data of the driving road of the vehicle and a preset target upper limit of the speed;
the following speed change difference value parameter is calculated according to the data of the vehicle, the data of the driving road of the vehicle, the data of the front vehicle closest to the vehicle and a preset target speed upper limit; and
and changing the difference parameter according to the speed of the difference between the target speed upper limit and the speed upper limit threshold value under the working condition under the condition that the vehicle is not in front and is calculated according to the data of the vehicle, the data of the driving road of the vehicle, the preset target speed upper limit and the speed upper limit threshold value of the vehicle.
Preferably, in the intelligent decision-making method for avoiding parallel driving applied to an automatic driving high-speed scene, the constraint conditions include:
estimating whether the vehicle can keep a safe distance with the front vehicle when the current running speed of the vehicle reaches the preset target speed upper limit under the working condition under the condition that the front vehicle exists according to the data of the vehicle, the data of the front vehicle and the preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; if not, optimizing the preset target speed upper limit;
estimating whether the time of the vehicle and the parallel driving vehicle in a staggered parallel driving area under the working condition is not more than the time of the vehicle overtaking a safe distance with the front vehicle or not under the condition that the front vehicle exists according to the data of the vehicle, the data of other vehicles and a preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; if not, optimizing the preset target speed upper limit; and
estimating whether the time of the parallel driving area staggered between the vehicle and the parallel driving vehicle under the working condition is not more than the time threshold value according to the data of the vehicle, the data of other vehicles and a preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; otherwise, optimizing the preset target speed upper limit.
Preferably, the intelligent decision method for avoiding parallel driving applied to the automatic driving high-speed scene is used, wherein the data of the other vehicle refers to the current driving speed of the other vehicle; the data of the front vehicle refers to the current running speed of the front vehicle; the data of the host vehicle indicates: the distance between the vehicle and the front vehicle, the current running speed of the vehicle, the upper limit threshold value of the speed for avoiding parallel driving and the upper limit value of the running speed set by the vehicle are adjusted; the data of the driving road of the vehicle refers to the speed limit value of the road.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the dynamic information includes: maximum and minimum limit speeds of the vehicle driving road, and information of a road section where the vehicle driving road is located; the current running speed of the vehicle, the current running acceleration of the vehicle and the maximum running speed set by the vehicle user; the number of vehicles around the host vehicle relative to the host vehicle, the current travel speed of the vehicles around the host vehicle, and the relative distance of the vehicles around the host vehicle from the sensors of the host vehicle.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the road section information includes: straight going sections, low-curvature turning sections, ramps, diverging sections, and converging sections.
Preferably, in the intelligent decision-making method for avoiding parallel driving applied to an automatic driving high-speed scene, the time threshold includes:
the time threshold value when the current running speed of the vehicle is higher than that of the parallel driving vehicle is set before the position of the vehicle relative to the parallel driving vehicle;
the position of the vehicle is forward relative to the parallel driving vehicle, and the current running speed of the vehicle is smaller than the time threshold value of the parallel driving vehicle;
the time threshold value when the current running speed of the vehicle is higher than the current running speed of the parallel driving vehicle is behind the position of the vehicle relative to the parallel driving vehicle; and
the position of the vehicle is back relative to the parallel driving vehicle, and the current running speed of the vehicle is smaller than the time threshold value of the parallel driving vehicle.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the parallel driving area refers to: the distance between the tail of the parallel-driving vehicle and the head of the vehicle is less than the distance between the head of the parallel-driving vehicle and the lane on which the vehicle runs and the safe cutting-in of other vehicles running in other lanes adjacent to the lane on which the vehicle runs, and the distance between the head of the parallel-driving vehicle and the tail of the vehicle is less than the distance between the safe cutting-in of the vehicle running behind the vehicle on the lane on which the vehicle runs and the lane on which the parallel-driving vehicle is located.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the adjusting of the driving speed plan of the vehicle in real time according to the target speed upper limit includes: the real-time pre-judgment is carried out on whether the vehicle and the parallel driving vehicle are still in a parallel driving situation or not when the time contained in the time threshold value is over; if so, optimizing the target speed upper limit of the vehicle through the evaluation function and the constraint condition again, and planning the running speed of the vehicle according to the new target speed upper limit; otherwise, the current upper speed limit is maintained until the vehicle is staggered from the parallel-driving vehicle.
Preferably, in the intelligent decision method for avoiding parallel driving applied to an automatic driving high-speed scene, the host vehicle and the parallel driving vehicle are staggered and then continue to keep the current driving speed for a preset time, and then the host vehicle gradually adjusts the upper speed limit to the upper speed limit set before the other vehicle closest to the host vehicle is judged to belong to the scene of parallel driving.
The invention at least comprises the following beneficial effects:
the invention relates to an intelligent decision method for avoiding parallel driving, which is applied to an automatic driving high-speed scene, and is characterized in that whether the vehicle belongs to a parallel driving situation with other vehicles is pre-judged in real time through data obtained by a sensor on the vehicle, then the upper limit of the speed required by the vehicle for staggering the parallel driving vehicle is calculated according to an evaluation function and a constraint condition, and finally the intelligent decision for avoiding the parallel driving is realized as flexibly as people through regulating the speed plan of the vehicle in advance, so that the situation that the vehicle is in parallel driving with other vehicles for a long time at a high speed is effectively avoided, namely, the situation that dynamic traffic jam occurs at the high speed is effectively avoided, and the riding experience of the automatic driving is also improved in a smooth high-.
The method adapts to a high-speed dynamic scene through real-time prejudgment and optimized calculation, and effectively improves the efficiency of staggered parallel driving of the automatic driving vehicle.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Detailed Description
The present invention is described in detail below to enable one skilled in the art to practice the invention in light of the description.
An intelligent decision-making method for avoiding parallel driving applied to an automatic driving high-speed scene comprises the following steps:
the method comprises the steps of acquiring dynamic information of a vehicle, a driving road of the vehicle and vehicles around the vehicle in real time, and pre-judging whether other vehicles closest to the vehicle belong to a scene of parallel driving according to the collected dynamic information;
if so, optimizing the target speed upper limit of the vehicle through an evaluation function and a constraint condition, and adjusting the driving speed plan of the vehicle in real time according to the target speed upper limit;
if not, continuing to run according to the current running speed plan of the vehicle;
the scene of prejudging whether other vehicles closest to the vehicle belong to parallel driving according to the collected dynamic information comprises the following steps:
and calculating the time value of other vehicles, which are closest to the vehicle, on the two lanes from the dynamic information and leave the parallel driving area of the vehicle, and when the time value exceeds a preset time threshold value of staggering the parallel driving area, prejudging the situation that the vehicle and the other vehicles are driven in parallel, and setting the other vehicles as parallel driving vehicles.
In the scheme, data of a road, a vehicle and surrounding vehicles collected in real time by a sensor on an automatic driving vehicle are used for calculating a time value required for the vehicles with two lanes closest to the vehicle to stagger a parallel driving area at the current relative speed, and when the estimated time value exceeds a preset time threshold value required for staggering the parallel driving area, the situation that the parallel driving situation occurs is predicted. At the moment, the optimized upper speed limit of the parallel driving area which can be safely and effectively staggered within a time threshold value which is less than the preset time threshold value can be optimized through evaluation functions and constraint conditions, and the speed plan of the vehicle is adjusted in real time, so that the estimation of the parallel driving situation which possibly occurs in the process of high-speed automatic driving, the intelligent upper speed limit decision and the real-time speed plan adjustment are realized, the occurrence of the dynamic traffic jam situation at high speed is effectively avoided, and the safety and experience of passengers are improved by not driving in parallel with a large vehicle or a dangerous vehicle for a long time.
In a preferred embodiment, the evaluation function includes: calculating a time parameter according to the data of the other vehicle and a preset target speed upper limit; calculating a speed difference value parameter for changing the upper limit of the speed according to the data of the vehicle, the data of the driving road of the vehicle and a preset target upper limit of the speed; the following speed change difference value parameter is calculated according to the data of the vehicle, the data of the driving road of the vehicle, the data of the front vehicle closest to the vehicle and a preset target speed upper limit; and changing a difference value parameter according to the speed of the difference between the target speed upper limit and the speed upper limit threshold value under the working condition under the condition that the vehicle is not provided with the front vehicle and calculated according to the data of the vehicle, the data of the driving road of the vehicle, a preset target speed upper limit and the speed upper limit threshold value of the vehicle.
In the above scheme, the time parameter calculated according to the data of the other vehicle and the preset target speed upper limit has the following meaning: the higher the upper limit of the target speed is, the smaller the time for staggering the parallel driving area is, namely the lower the corresponding cost is, otherwise, the higher the cost is; the significance of the speed difference value parameter of the change speed upper limit calculated according to the data of the vehicle, the data of the driving road and the preset target speed upper limit is as follows: the smaller the difference between the preset target speed upper limit and the original speed upper limit of the vehicle is, the smaller the cost is, and otherwise, the higher the cost is, wherein the original speed upper limit of the vehicle refers to the minimum speed upper limit obtained after the speed upper limit set by a vehicle user is compared with the road speed upper limit; the significance of the following speed change difference parameter calculated according to the data of the vehicle, the data of the driving road, the data of the front vehicle closest to the vehicle and the preset target speed upper limit is as follows: when the lane has detected front vehicles and the current speed of the front vehicles is lower than the upper limit threshold value of the speed of the vehicle, the closer the preset target upper limit speed is to the current speed of the front vehicle closest to the vehicle, the lower the cost is, otherwise, the higher the cost is; according to the data of the vehicle, the data of the driving road, the preset target speed upper limit and the speed upper limit threshold of the vehicle, under the working condition that no vehicle exists or the current speed of the vehicle is higher than the speed upper limit threshold of the vehicle, the significance of the speed change difference parameter of the difference between the preset target speed upper limit of the vehicle and the speed upper limit threshold of the vehicle is as follows: under the working condition, the cost is smaller as the preset target speed upper limit is closer to the speed upper limit threshold, otherwise, the cost is larger, namely under the working condition that no front vehicle exists or the current speed of the front vehicle is higher than the speed upper limit threshold of the vehicle, the vehicle is more prone to accelerating as much as possible and staggering with a parallel driving vehicle.
By introducing speed and time into the evaluation function, the final optimized speed upper limit can not be unconditionally taken and can not be meaninglessly maintained, so that a result obtained after speed and time are balanced is output, and the intelligence and universality of the decision method are embodied.
In a preferred embodiment, the constraint condition includes: estimating whether the vehicle can keep a safe distance with the front vehicle when the current running speed of the vehicle reaches the preset target speed upper limit under the working condition under the condition that the front vehicle exists according to the data of the vehicle, the data of the front vehicle and the preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; if not, optimizing the preset target speed upper limit; estimating whether the time of the vehicle and the parallel driving vehicle in a staggered parallel driving area under the working condition is not more than the time of the vehicle overtaking a safe distance with the front vehicle or not under the condition that the front vehicle exists according to the data of the vehicle, the data of other vehicles and a preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; if not, optimizing the preset target speed upper limit; whether the time of the parallel driving area where the vehicle and the parallel driving vehicle are staggered under the working condition is not more than the time threshold value or not is estimated according to the data of the vehicle, the data of other vehicles and the preset target speed upper limit; if yes, the preset target speed upper limit meets the constraint; otherwise, optimizing the preset target speed upper limit.
In the scheme, the data of the front vehicle and the time threshold are introduced into the constraint condition, so that the speed upper limit which cannot meet the time threshold is eliminated, and the speed upper limit which cannot be realized due to the existence of the front vehicle of the current lane is met, the range of the feasible threshold is narrowed, the reliability and the safety of the output speed upper limit are ensured, and the calculation efficiency is also improved.
In a preferred embodiment, the other vehicle data indicates a current driving speed of the other vehicle; the data of the front vehicle refers to the current running speed of the front vehicle; the data of the host vehicle indicates: the distance between the vehicle and the front vehicle, the current running speed of the vehicle, the upper limit threshold value of the speed for avoiding parallel driving and the upper limit value of the running speed set by the vehicle are adjusted; the data of the driving road of the vehicle refers to the speed limit value of the road.
In the above aspect, adjusting the upper speed limit threshold for avoiding parallel driving includes: the maximum tolerance value of the current upper speed limit is increased, the maximum tolerance value of the current upper speed limit is reduced, and the maximum plus-minus tolerance value of the target upper speed limit is defined and adjusted, so that the upper speed limit optimally output through an evaluation function and a constraint condition is not deviated from the upper speed limit before parallel driving decision making, the situation that the speed change of the vehicle is too large is avoided, the purpose of staggering a parallel area with the parallel vehicle is achieved, and the experience of passengers is not influenced.
In a preferred embodiment, the dynamic information includes: maximum and minimum limit speeds of the vehicle driving road, and information of a road section where the vehicle driving road is located; the current running speed of the vehicle, the current running acceleration of the vehicle and the maximum running speed set by the vehicle user; the number of vehicles around the host vehicle relative to the host vehicle, the current travel speed of the vehicles around the host vehicle, and the relative distance of the vehicles around the host vehicle from the sensors of the host vehicle.
In the above-mentioned solution, by collecting data of the vehicle driving road, the vehicle, and the vehicles around the vehicle in detail, comprehensive information is provided for judging the parallel driving scenario, so that the accuracy of predicting the parallel driving scenario is improved.
In a preferred embodiment, the road section information includes: straight going sections, low-curvature turning sections, ramps, diverging sections, and converging sections.
In the scheme, the road section suitable for the intelligent decision method is definitely specified by defining the road section type, the possibility that the driving danger occurs in the road sections such as a curve with large curvature, a toll station entrance and the like due to speed regulation planning is eliminated, and the safety and the reliability of the intelligent decision method are improved.
In a preferred embodiment, the time threshold includes: the time threshold value when the current running speed of the vehicle is higher than that of the parallel driving vehicle is set before the position of the vehicle relative to the parallel driving vehicle; the position of the vehicle is forward relative to the parallel driving vehicle, and the current running speed of the vehicle is smaller than the time threshold value of the parallel driving vehicle; the time threshold value when the current running speed of the vehicle is higher than the current running speed of the parallel driving vehicle is behind the position of the vehicle relative to the parallel driving vehicle; and a time threshold value when the current running speed of the vehicle is lower than the current running speed of the parallel driving vehicle.
In the scheme, the four scenes are divided according to the relative positions and the running speeds of the vehicle and the parallel driving vehicle, corresponding time thresholds are respectively set for the scenes, and the judgment condition of the parallel driving is refined, so that the reliability of the intelligent decision method is improved.
In a preferred embodiment, the parallel driving area refers to: the distance between the tail of the parallel-driving vehicle and the head of the vehicle is less than the distance between the head of the parallel-driving vehicle and the lane on which the vehicle runs and the safe cutting-in of other vehicles running in other lanes adjacent to the lane on which the vehicle runs, and the distance between the head of the parallel-driving vehicle and the tail of the vehicle is less than the distance between the safe cutting-in of the vehicle running behind the vehicle on the lane on which the vehicle runs and the lane on which the parallel-driving vehicle is located.
In the scheme, the parallel driving area is defined by considering the safe distance between the front and the rear of the vehicle, so that the situation that other vehicles on two lanes can safely cut into the vehicle lane is ensured, and the vehicles behind the vehicle lane can safely merge from the vehicle lane to the two lanes on two sides, thereby avoiding the high-speed dynamic traffic jam situation, promoting good traffic environment and improving the automatic driving experience of passengers.
In a preferred embodiment, the adjusting the driving speed plan of the vehicle in real time according to the target upper speed limit includes: the real-time pre-judgment is carried out on whether the vehicle and the parallel driving vehicle are still in a parallel driving situation or not when the time contained in the time threshold value is over; if so, optimizing the target speed upper limit of the vehicle through the evaluation function and the constraint condition again, and planning the running speed of the vehicle according to the new target speed upper limit; otherwise, the current upper speed limit is maintained until the vehicle is staggered from the parallel-driving vehicle.
In the scheme, the optimal target speed upper limit driving can be adopted to realize the staggering between the vehicle and the parallel driving vehicle within the limited time threshold value through real-time prejudgment and adjustment, so that the target speed upper limit of the vehicle can be timely optimized and adjusted when the vehicle and the parallel driving vehicle cannot be staggered within the time threshold value through prejudgment, the timely staggering between the vehicle and the parallel driving vehicle is facilitated, namely, the corresponding real-time decision on the change of a dynamic environment is realized, the adaptability to the dynamic environment is improved, and the humanization and the intellectualization of the intelligent decision method are embodied.
In a preferred scheme, after the vehicle and the parallel driving vehicle are staggered, the vehicle continues to keep the current running speed for a preset time, and then the vehicle gradually adjusts the upper speed limit to the upper speed limit set before the other vehicle closest to the vehicle is judged to belong to the scene of parallel driving.
In the above scheme, after the vehicle and the parallel-driving vehicle are separated from each other by the parallel area, the vehicle continues to keep running at the current running speed for a preset time, that is, the vehicle runs for a period of time at the current running speed, so that the possibility that the vehicle and the parallel-driving vehicle enter the parallel-driving scene again can be greatly reduced. Wherein, according to the repeated drive test, the time is set to 3-10 seconds, which can meet the requirement of greatly reducing the possibility that the same vehicle and the same vehicle enter the parallel driving scene repeatedly.
While embodiments of the invention have been disclosed above, it is not limited to the applications listed in the description and the embodiments, which are fully applicable in all kinds of fields of application of the invention, and further modifications may readily be effected by those skilled in the art, so that the invention is not limited to the specific details without departing from the general concept defined by the claims and the scope of equivalents.