CN105760822A

CN105760822A - Vehicle driving control method and system

Info

Publication number: CN105760822A
Application number: CN201610070331.1A
Authority: CN
Inventors: 徐黎卿
Original assignee: Individual
Current assignee: Individual
Priority date: 2016-02-01
Filing date: 2016-02-01
Publication date: 2016-07-13
Anticipated expiration: 2036-02-01
Also published as: CN105760822B

Abstract

The present invention provides a vehicle driving control method and system. The method comprises: performing real-time tracking and collection of driver's two-hand multi-frame first image information when driver's two-hand information is detected in a preset tracking region; analyzing the multi-frame first image information to obtain the driver's two-hand first running track; tracking and collecting multi-frame second image information of a vehicle steering wheel in the preset tracking region in real time; analyzing the multi-frame image information to obtain the second running track of the vehicle steering wheel; controlling the vehicle to realize mode switching according to the matching degree of the first running track and the second running track, at the same time, extracting the track information of the driver's hands and the vehicle steering wheel, and applying the driver intention corresponding to the interaction of the track information of the driver's hands and the vehicle steering wheel to the vehicle to control the vehicle, so that the switching safety of the driving modes is improved, the flexible switching and harmonious interaction of automatic driving and manual driving are realized, and the practicability is high.

Description

A kind of vehicle drive control method and system

Technical field

The present invention relates to vehicle drive and control technical field, in particular to a kind of vehicle drive control method and system.

Background technology

Along with vehicle intellectualized development, automatic Pilot technology is increasingly subject to the attention of vast automobile manufacturing company.Using in advance as coml, ADAS (AdvancedDriverAssistantSystem, senior drive assist system) technology has been obtained for sufficient development at present, and has had the example of many application.Such as, a lot of automobiles have automatic cruising, automatic parking, the application of the correlation technique of the functions such as track maintenance.

Automatic Pilot technology is very harsh for the requirement of safety, and the switching between itself and pilot steering is with greater need for complete Security Techniques.Above-mentioned switching is two-way, namely includes: pilot steering pattern switches to automatic driving mode and automatic driving mode to switch to pilot steering pattern.From pilot steering to automatic Pilot switch time, automobile carries out the intention of automatic Pilot to carry out corresponding pattern switching firstly the need of obtaining driver's expectation, wherein, above-mentioned intention can press some button by some and be achieved, it is also possible to by being achieved after both hands off-direction dish some seconds simultaneously；And from automatic Pilot to pilot steering switch time, more increasingly complex than the former switching, particularly when some is urgent, such as, driver find driving locus existing problems and have carry out pilot steering immediately intention to carry out corresponding pattern switching, visible, how to ensure the intention of driver can be sensed by automobile and can obtain further with discharge automobile manipulation power become a problem demanding prompt solution.

Consider that driver discharges and obtains the ability of manipulation power and should correspond to the switching of above-mentioned pattern, and steering is as the important step in automotive control system, correspond to again driver's release and obtains the ability of manipulation power.At present, by the steering product E PS (ElectricalPowerSteering of main flow, electronic power assist steering system) carry out the control of automobile, utilize the torque sensor in EPS can obtain whether driver applies hand-power state on the steering wheel to carry out the judgement of driver intention.But, rely on torque sensor as the intention of driver, be not enough to the switching by above-mentioned state control model.Such as, when torque sensor senses that nothing applying hand-power is on the steering wheel, automobile is switched to automatic Pilot from pilot steering, but produce above-mentioned to be probably during craspedodrome steering wheel without rotating without torsional forces without the state applying hand-power, do not represent driver's expectation and carry out above-mentioned switching, in addition, if the switching of pilot steering and automatic Pilot is indefinite, when pilot steering misjudgement is for automatic Pilot, itself there is potential safety hazard, and the behavior of both hands off-direction dish is also that law is clearly forbidden when pilot steering.

Inventor finds under study for action, the mode adopting torque sensor in prior art carries out the control of mode motor switching, the either-or cold cut of above-mentioned switching regulator is changed, security reliability is poor, and how flexible switching is carried out for automatic Pilot and pilot steering, harmonious mutual control system there is no perfect scheme.

Summary of the invention

In view of this, it is an object of the invention to provide a kind of vehicle drive control method and system, information between being intended to by steering wheel for vehicle and driver's both hands is mutual, it is achieved that the flexible switching of automatic Pilot and pilot steering, it is ensured that the safety in handoff procedure.

First aspect, embodiments provides a kind of vehicle drive control method, and the method specifically includes:

When driver's both hands information being detected in default tracing area, real-time tracking gathers driver's both hands multiframe the first image information in default tracing area；

It is analyzed processing to multiframe the first image information gathered, obtains the first running orbit of driver's both hands；

Real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area；

It is analyzed processing to multiframe the second image information gathered, obtains the second running orbit of steering wheel for vehicle；

Switch mode corresponding to matching degree according to the first running orbit and the second running orbit controls the switching of vehicle implementation pattern；Wherein, pattern includes automatic driving mode and pilot steering pattern.

In conjunction with first aspect, embodiments providing the first possible embodiment of first aspect, wherein, real-time tracking gathers driver's both hands multiframe the first image information in default tracing area and includes:

Gather driver's both hands first frame the first image information in default tracing area；

When driver's both hands first frame the first image information in default tracing area being detected, extract the fisrt feature information in the first image information；Fisrt feature information includes the following information of driver's both hands: the first color characteristic information and the first body dynamics information；First body dynamics information at least includes the primary importance point of driver's both hands；

The fisrt feature information of first frame the first image information is calculated processing by the particle filter algorithm according to multiple characteristics space, obtains the primary importance point of next frame the first image information；

When often receiving frame first image information, extract the fisrt feature information in the first image information, and fisrt feature information is calculated processing by the particle filter algorithm according to multiple characteristics space, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that the first image information is corresponding.

The first possible embodiment in conjunction with first aspect, embodiments provide the embodiment that the second of first aspect is possible, wherein, the fisrt feature information of first frame the first image information is calculated processing by the particle filter algorithm according to multiple characteristics space, and the primary importance point obtaining next frame the first image information includes:

Multiple target search particle meeting Gauss distribution is set in the preset range region of next frame the first image information；Wherein, the search graph picture that each target search particle is corresponding different respectively, search graph picture is the image of zones of different in next frame the first image information；

Extract the second feature information of each search graph picture；Second feature information includes the following information of search graph picture: the second color characteristic information and the second body dynamics information；Second body dynamics information at least includes the second position point of search graph picture；

First color characteristic information and the second color characteristic information are carried out color similarity calculating, obtains the second color characteristic information the highest with the first color characteristic information similarity；

The second position point of search graph picture corresponding for the second color characteristic information obtained is defined as the primary importance point of driver's both hands in next frame the first image information.

In conjunction with the embodiment that the second of first aspect is possible, embodiments providing the third possible embodiment of first aspect, wherein, real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area includes:

The collection vehicle steering wheel first frame the second image information in default tracing area；

In first frame the second image information, calculate the absolute coordinate of the desired guiding trajectory starting point of steering wheel for vehicle；

It is calculated processing to the absolute coordinate of the desired guiding trajectory starting point in first frame the second image information according to Kalman filtering algorithm, obtains the absolute coordinate of next frame the second image information；

When often receiving frame second image information, extract the absolute coordinate of desired guiding trajectory starting point in the second image information, obtain the absolute coordinate of desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the desired guiding trajectory starting point in all of second image information of output collection and the second image information.

The third possible embodiment in conjunction with first aspect, embodiments provide the 4th kind of possible embodiment of first aspect, wherein, control the switching of vehicle implementation pattern according to the switch mode that the matching degree of the first running orbit and the second running orbit is corresponding to include:

The first matching degree according to the first running orbit and the second running orbit, generates first driving model switching signal corresponding with the first matching degree；First driving model switching signal is used for controlling vehicle and is switched to pilot steering pattern by automatic driving mode；

The second matching degree according to the first running orbit and the second running orbit, generates second driving model switching signal corresponding with the second matching degree；Second driving model switching signal is used for controlling vehicle and is switched to automatic driving mode by pilot steering pattern.

Second aspect, the embodiment of the present invention additionally provides a kind of vehicle drive and controls system, and this system includes:

First image information follows the tracks of acquisition module, and during for driver's both hands information being detected in default tracing area, real-time tracking gathers driver's both hands multiframe the first image information in default tracing area；

First analysis and processing module, multiframe the first image information for the first image information is followed the tracks of acquisition module collection is analyzed processing, and obtains the first running orbit of driver's both hands；

Second image information follows the tracks of acquisition module, for real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area；

Second analysis and processing module, multiframe the second image information for the second image information is followed the tracks of acquisition module collection is analyzed processing, and obtains the second running orbit of steering wheel for vehicle；

Pattern switching control module, controls the switching of vehicle implementation pattern for the switch mode processing, according to the first analysis and processing module, the first running orbit of obtaining and the second analysis and processing module processes the matching degree of the second running orbit obtained corresponding；Wherein, pattern includes automatic driving mode and pilot steering pattern.

In conjunction with second aspect, embodiments providing the first possible embodiment of second aspect, wherein, the first image information is followed the tracks of acquisition module and is included:

First image information collecting unit, for gathering driver's both hands first frame the first image information in default tracing area；

Fisrt feature information extraction unit, for when the first image information collecting unit collects first frame the first image information, extracting the fisrt feature information in the first image information；Fisrt feature information includes the following information of driver's both hands: the first color characteristic information and the first body dynamics information；First body dynamics information at least includes the primary importance point of driver's both hands；

Location point computing unit, for being calculated processing to the fisrt feature information that fisrt feature information extraction unit is extracted according to the particle filter algorithm in multiple characteristics space, obtains the primary importance point of next frame the first image information；

Fisrt feature information extraction unit is additionally operable to, and when often receiving frame first image information, extracts the fisrt feature information in the first image information；Location point computing unit is additionally operable to, the fisrt feature information that fisrt feature information extraction unit is extracted is calculated processing by the particle filter algorithm according to multiple characteristics space, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that the first image information is corresponding.

In conjunction with the first possible embodiment of second aspect, embodiments providing the embodiment that the second of second aspect is possible, wherein, location point computing unit includes:

Target search particle arranges subelement, for arranging multiple target search particle meeting Gauss distribution in the preset range region of next frame the first image information；Wherein, the search graph picture that each target search particle is corresponding different respectively, search graph picture is the image of zones of different in next frame the first image information；

Second feature information retrieval subelement, for extracting the second feature information of each search graph picture；Second feature information includes the following information of search graph picture: the second color characteristic information and the second body dynamics information；Second body dynamics information at least includes the second position point of search graph picture；

Color similarity computation subunit, the the second color characteristic information extracted for the first color characteristic information that fisrt feature information extraction unit is extracted and second feature information retrieval subelement carries out color similarity calculating, obtains the second color characteristic information the highest with the first color characteristic information similarity；

Location point computation subunit, is defined as the primary importance point of driver's both hands in next frame the first image information for the second position point of search graph picture corresponding to the second color characteristic information of color similarity computation subunit being obtained.

In conjunction with the embodiment that the second of second aspect is possible, embodiments providing the third possible embodiment of second aspect, wherein, the second image information is followed the tracks of acquisition module and is included:

Second image information collecting unit, for the collection vehicle steering wheel first frame the second image information in default tracing area；

First absolute coordinate computing unit, for, in first frame the second image information that the second image information collecting unit gathers, calculating the absolute coordinate of the desired guiding trajectory starting point of steering wheel for vehicle；

Second absolute coordinate computing unit, for being calculated processing to the absolute coordinate that the first absolute coordinate computing unit calculates according to Kalman filtering algorithm, obtains the absolute coordinate of the desired guiding trajectory starting point of next frame the second image information；

First absolute coordinate computing unit is additionally operable to, and when often receiving frame second image information, extracts the absolute coordinate of desired guiding trajectory starting point in the second image information；Second absolute coordinate computing unit is additionally operable to, process according to the absolute coordinate that the first absolute coordinate computing unit calculates, obtain the absolute coordinate of desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the desired guiding trajectory starting point in all of second image information of output collection and the second image information.

In conjunction with the third possible embodiment of second aspect, embodiments providing the 4th kind of possible embodiment of second aspect, wherein pattern switching control module includes:

First switching signal generates unit, process the first matching degree of the second running orbit obtained for the first running orbit obtained according to the first analysis and processing module process and the second analysis and processing module, generate first driving model switching signal corresponding with the first matching degree；First driving model switching signal is used for controlling vehicle and is switched to pilot steering pattern by automatic driving mode；

Second switching signal generates unit, process the second matching degree of the second running orbit obtained for the first running orbit obtained according to the first analysis and processing module process and the second analysis and processing module, generate second driving model switching signal corresponding with the second matching degree；Second driving model switching signal is used for controlling vehicle and is switched to automatic driving mode by pilot steering pattern.

The vehicle drive control method of embodiment of the present invention offer and system, first real-time tracking gathers driver's both hands and steering wheel for vehicle multiframe the first image information corresponding in default tracing area and multiframe the second image information, and is calculated above-mentioned multiframe the first image information and multiframe the second image information respectively processing the first running orbit obtaining correspondence and the second running orbit；Then the switching of vehicle implementation pattern is controlled according to the matching degree of the first running orbit and the second running orbit, with the control mode that employing torque sensor of the prior art carries out mode motor switching, security reliability is poor to be compared, it extracts the trace information of driver's hand and vehicle steering simultaneously, and can by the mutual corresponding driver intention of both trace informations for being applied on vehicle and this vehicle being controlled, the switch safety making driving model is higher, and achieve the flexible switching of automatic Pilot and pilot steering and harmonious mutual, also make practicality stronger.

For making the above-mentioned purpose of the present invention, feature and advantage to become apparent, preferred embodiment cited below particularly, and coordinate appended accompanying drawing, it is described in detail below.

Accompanying drawing explanation

In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, the accompanying drawing used required in embodiment will be briefly described below, it is to be understood that, the following drawings illustrate only certain embodiments of the present invention, therefore the restriction to scope it is not construed as, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other relevant accompanying drawings according to these accompanying drawings.

Fig. 1 illustrates the method flow diagram that a kind of vehicle drive that the embodiment of the present invention provides controls；

Fig. 2 illustrates the method flow diagram that the another kind of vehicle drive that the embodiment of the present invention provides controls；

Fig. 3 illustrates the method flow diagram that the another kind of vehicle drive that the embodiment of the present invention provides controls；

Fig. 4 illustrates the method flow diagram that the another kind of vehicle drive that the embodiment of the present invention provides controls；

Fig. 5 illustrates the method flow diagram that the another kind of vehicle drive that the embodiment of the present invention provides controls；

Fig. 6 illustrates that a kind of vehicle drive that the embodiment of the present invention provides controls the structural representation of system；

Fig. 7 illustrates that in a kind of vehicle drive control system that the embodiment of the present invention provides, the first image information follows the tracks of the structural representation of acquisition module；

Fig. 8 illustrates the structural representation of location point computing unit in a kind of vehicle drive control system that the embodiment of the present invention provides；

Fig. 9 illustrates that in a kind of vehicle drive control system that the embodiment of the present invention provides, the second image information follows the tracks of the structural representation of acquisition module；

Figure 10 illustrates the structural representation of pattern switching control module in a kind of vehicle drive control system that the embodiment of the present invention provides.

Main element symbol description:

11, the first image information follows the tracks of acquisition module；22, the first analysis and processing module；33, the second image information follows the tracks of acquisition module；44, the second analysis and processing module；55, pattern switching control module；111, the first image information collecting unit；112, fisrt feature information extraction unit；113, location point computing unit；331, the second image information collecting unit；332, the first absolute coordinate computing unit；333, the second absolute coordinate computing unit；551, the first switching signal generates unit；552, the second switching signal generates unit；1131, target search particle arranges subelement；1132, second feature information retrieval subelement；1133, color similarity computation subunit；1134, location point computation subunit.

Detailed description of the invention

Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Generally can with various different configurations arrange and design with the assembly of the embodiment of the present invention that illustrate described in accompanying drawing herein.Therefore, below the detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit claimed the scope of the present invention, but is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, the every other embodiment that those skilled in the art obtain under the premise not making creative work, broadly fall into the scope of protection of the invention.

Consider correlation technique to adopt torque sensor carry out the control mode of mode motor switching, security reliability is poor and cannot be carried out the problems such as the flexible switching between automatic Pilot and pilot steering, harmonious mutual and practicality be poor, embodiments provide a kind of vehicle drive control method and system, driver intention can be applied to wagon control itself, security reliability is higher, and achieving the flexibility switching of automatic Pilot and pilot steering, harmony alternately, practicality is stronger.The flow chart of the vehicle drive control method that the embodiment of the present invention shown in Figure 1 provides, described method specifically includes following steps:

S101, when driver's both hands information being detected in default tracing area, real-time tracking gathers driver's both hands multiframe the first image information in default tracing area；

S102, to gather multiframe the first image information be analyzed process, obtain the first running orbit of driver's both hands.

Concrete, consider in the process judging driver's driving intention, core point is the track algorithm of driver's both hands the most, when the vehicle drive control method that the embodiment of the present invention provides detects driver's both hands information in default tracing area, real-time tracking gathers described driver's both hands multiframe the first image information in default tracing area, this first image information can include the histogram frame of driver's both hands, it is also possible to is the irregular image including driver's both hands profile；Multiframe first image information of collection is carried out extraction and the combination of the tracing point of correspondence, obtains the first running orbit of driver's both hands.

Wherein, above-mentioned default tracing area is the region comprising steering wheel for vehicle, and when driving both hands and steering wheel for vehicle has mutual, driver's both hands are also in this default tracing area；When driver's both hands and steering wheel for vehicle no interactions, driver's both hands leave above-mentioned default tracing area.Wherein, preset image corresponding to tracing area and can carry out Real-time Collection by the monitored video camera system being loaded into above vehicle drive district, initialized image procossing is carried out based on these images, extract initialization information corresponding when both hands occur first, and pass through, according to this initialization information, the tracking carrying out driver's both hands that particle filter is real-time, obtain the positional information following the tracks of multiframe the first image information corresponding to each tracing point gathering both hands and this tracing point to obtain the first running orbit corresponding to driver's both hands.Wherein, the tracking of driver's both hands can be the tracking respectively that driver is singlehanded, it is also possible to follow the tracks of the while of being driver's both hands.

S103, real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area；

S104, to gather multiframe the second image information be analyzed process, obtain the second running orbit of steering wheel for vehicle.

Concrete, in order to better carry out the judgement of driver's driving intention, the vehicle drive control method that the embodiment of the present invention provides also real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area, the extraction of corresponding absolute coordinate is carried out, to obtain the second running orbit of corresponding this steering wheel for vehicle with each absolute coordinate by the track initial point preset in above-mentioned second image information.Wherein, above-mentioned steering wheel for vehicle comes across in the monitored video camera system corresponding to above-mentioned default tracing area all the time.

S105, control the switching of vehicle implementation pattern according to the switch mode that the matching degree of the first running orbit and the second running orbit is corresponding；Wherein, pattern includes automatic driving mode and pilot steering pattern.

Concrete, the vehicle drive control method that the embodiment of the present invention provides controls the pattern switching of vehicle by the matching degree of above-mentioned first running orbit and the second running orbit.

Wherein, consider that the second running orbit of steering wheel for vehicle exists in whole vehicle drive controls always, and the first running orbit of driver's both hands only just exists when driver has Driving control to be intended to, therefore, when carrying out the switching of automatic driving mode extremely artificial driving model, above-mentioned matching degree is the matching initial moment as reference in the initial track point moment that the driver's both hands to extract enter monitored video camera system, now, control vehicle by the switch mode corresponding to the matching degree of the first running orbit and the second running orbit to realize switching to pilot steering pattern from automatic driving mode；When carrying out pilot steering pattern to the switching of automatic driving mode, above-mentioned matching degree is to leave the matching initial moment as reference in the final tracing point moment of monitored video camera system using the driver's both hands extracted, now, control vehicle by the switch mode corresponding to the matching degree of the first running orbit and the second running orbit and realize the switching from pilot steering pattern to automatic driving mode.

The vehicle drive control method that the embodiment of the present invention provides, first real-time tracking gathers driver's both hands and steering wheel for vehicle multiframe the first image information corresponding in default tracing area and multiframe the second image information, and is calculated above-mentioned multiframe the first image information and multiframe the second image information respectively processing the first running orbit obtaining correspondence and the second running orbit；Then the switching of vehicle implementation pattern is controlled according to the matching degree of the first running orbit and the second running orbit, with the control mode that employing torque sensor of the prior art carries out mode motor switching, security reliability is poor to be compared, it extracts the trace information of driver's hand and vehicle steering simultaneously, and can by the mutual corresponding driver intention of both trace informations for being applied on vehicle and this vehicle being controlled, the switch safety making driving model is higher, and achieve the flexible switching of automatic Pilot and pilot steering and harmonious mutual, also make practicality stronger.

In order to better realize the trajectory extraction to driver's both hands, the tracking gatherer process of the invention described above S101, realize especially by following steps, flow chart shown in Figure 2, described method also includes:

S201, collection driver's both hands first frame the first image information in default tracing area；

S202, when driver's both hands first frame the first image information in default tracing area being detected, extract the fisrt feature information in the first image information；Wherein, fisrt feature information includes the following information of driver's both hands: the first color characteristic information and the first body dynamics information；First body dynamics information at least includes the primary importance point of driver's both hands.

Concrete, when driver's both hands occur in the default tracing area that monitored video camera system is corresponding, the vehicle drive control method that the embodiment of the present invention provides all can accurately be followed the tracks of and be gathered the first image information that driver's both hands are corresponding.

Wherein, monitored video camera system is when capturing driver's both hands first, gather first frame the first image information of current drivers's both hands, driver's both hands can be carried out segmentation by methods such as pattern recognitions and identify by above-mentioned gatherer process, it is also possible to by driver's both hands of moving object segmentation collection campaign.It is analyzed processing to above-mentioned first frame the first image information by feature extraction mode, obtains the fisrt feature information corresponding to the first image information.

In order to improve the robustness that follow-up tracking processes, the fisrt feature information of the first image information extracted in the vehicle drive control method that the embodiment of the present invention provides is the double characteristic information of driver's both hands, this double characteristic information includes the first color characteristic information and body dynamics information, wherein, this first color characteristic information is based on the image feature space of rgb space and based on the features of skin colors space of HSV space；Above-mentioned body dynamics information is based on the motion feature space of front and back frame, and at least includes the location point of driver's both hands.

Concrete, in the image of monitored video camera system real time shooting, for the extraction of the first color characteristic information, after being converted into HSV space by the rgb space of original image, re-use HSV feature space and carry out threshold values judgement, thus can more effectively extract the Skin Color Information of driver's both hands, again through using BC (Bhattacharyya) coefficient to obtain the probability distribution of hand central point, wherein high probability region is driver's both hands and low probability region is preset in tracing area other parts outside driver's both hands；For motion feature space, the information (based on gray space) of two frames carry out mathematic interpolation and obtain motion feature space before and after extracting, gray scale difference value corresponding to motion feature space can extract the information in the region of all high differences by corresponding algorithm, including the hand exercise probability distribution of steer direction dish.Owing to there is hand exercise, therefore there is high probability distribution at both hands place, and only exist low probability distribution at static place.

In addition, this fisrt feature information is to gather the first color characteristic information and body dynamics information and reassign the total information obtained through features described above information weight, wherein, this total information can be a concrete characteristic vector, and the robust tracking that this total information is follow-up processes the condition that provides the foundation.

S203, it is calculated processing to the fisrt feature information of first frame the first image information according to the particle filter algorithm in multiple characteristics space, obtains the primary importance point of next frame the first image information.

Consider that driver's both hands are as compliant motion target, in carrying out the process of image procossing being correlated with, being highly susceptible to the restriction of soft objectives self character and the situations such as loss occur to follow the tracks of, the vehicle drive control method that the embodiment of the present invention provides preferably adopts the tracking that the particle filter algorithm based on above-mentioned double characteristic information carries out follow-up driver's both hands to process.From the detailed process of the primary importance point of fisrt feature information prediction next frame first image information of current first frame the first image information, referring to Fig. 3.

S301, multiple target search particle meeting Gauss distribution is set in the preset range region of next frame the first image information；Wherein, the search graph picture that each target search particle is corresponding different respectively, search graph picture is the image of zones of different in next frame the first image information；

S302, extract the second feature information of each search graph picture；Second feature information includes the following information of search graph picture: the second color characteristic information and the second body dynamics information；Second body dynamics information at least includes the second position point of search graph picture；

S303, the first color characteristic information and the second color characteristic information are carried out color similarity calculating, obtain the second color characteristic information the highest with the first color characteristic information similarity；

S304, the second position point of search graph picture corresponding to the second color characteristic information obtained is defined as the primary importance point of driver's both hands in next frame the first image information.

Consider that traditional particle filter is predicted mainly by Gauss at forecast period, or Gauss prediction adds translational speed and judges, although these Forecasting Methodology amounts of calculation are little, but often occur in a practical situation to follow the tracks of mistake.The vehicle drive control method that the embodiment of the present invention provides carries out corresponding precomputation based on double characteristic information, thus improving the more intended particle of forecast period to be positioned at high probability region.

Concrete, compare and target search particle is set uniformly and causes subsequent characteristics to extract process consuming time longer at the whole plane of delineation that whole default tracing area is corresponding, first the present invention arranges multiple target search particle meeting Gauss distribution in the preset range region of next frame the first image information, namely it is that target search particles are being set near the more local of target (i.e. driver's both hands) more, some target search particles are set less at wide place, wherein, the search graph picture that each target search particle is corresponding different respectively, search graph picture is the image of zones of different in next frame the first image information, then the second feature information of each search graph picture is extracted；Same, above-mentioned second feature information includes the second color characteristic information and second body dynamics information of search graph picture, and wherein, the second body dynamics information at least includes the second position point of search graph picture；The Similarity Measure of first color histogram (corresponding first color characteristic information) of first frame the first image information then obtained according to initial phase and second color histogram (corresponding second color characteristic information) of target search image, obtain second color histogram the highest with the first color histogram similarity, finally also to arrange at the target search particle place that similarity highest point is corresponding and more search for particle, the target search particle place corresponding in similarity lowest point arranges less search particle, continue to calculate similarity, iterative cycles, the primary importance point of driver's both hands in next frame the first image information is achieved by stochastical sampling and importance resampling, namely a step dynamic tracking of driver's both hands is completed.

S204, when often receiving frame first image information, extract the fisrt feature information in the first image information, and fisrt feature information is calculated processing by the particle filter algorithm according to multiple characteristics space, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that the first image information is corresponding.

nullConcrete，The lasting real-time tracking in the whole process that driver's both hands occur monitored video camera system can be realized by endless form based on the above-mentioned detailed process from the primary importance point of fisrt feature information prediction next frame first image information of current first frame the first image information，After adding the acceptable precomputation of part，Search particle is by the more effective pixel occurring in more high probability，Compare after comparatively fast thus avoiding some hand motion，Particle filter can lose the target of tracking，It is by the position of the correlated characteristic information real-time update target of target search particle，Until when collecting last frame the first image information，Export all of first image information of collection and the primary importance point that this first image information is corresponding，Consider the irregular phenomenon of profile of both hands，The embodiment of the present invention preferably chooses the center of driver's both hands as location point corresponding to the first image information.

Additionally, driver's both hands are as soft objectives, its attitude, profile etc. are in dynamic change, the illumination variation that moment exists in the scene, bring challenges also can to the tracking of driver's both hands, it is also contemplated that when judging the driving intention of driver, most crucial function is the track algorithm of driver's both hands, the embodiment of the present invention is preferably chosen multiple characteristics space and is carried out particle filter tracking, additionally it is possible to bring following effect:

1, the tracing point of both hands can accurately be obtained, and when ensureing the credibility of tracing point, it is possible to obtain state description by the comparison with the arc point of steering wheel；

2, possessing both hands robust tracking, the track algorithm robustness that the embodiment of the present invention provides is relatively strong, in the situation that both hands exist close to and cover, and also can without the lasting tracking lost；

3, possessing the function again detecting both hands, when both hands are from after disappearing in image pickup scope, the embodiment of the present invention possesses the function again obtaining two-hand positions.

To sum up, after extracting the first image information and primary importance point thereof, track combination is passed through, it is possible to be easily obtained the first running orbit corresponding to driver's both hands.

Information in order to better realize driver's both hands and steering wheel for vehicle is mutual, the vehicle drive control method that the embodiment of the present invention provides also includes the extraction for the second running orbit point corresponding to the absolute coordinate of desired guiding trajectory starting point on steering wheel for vehicle, referring to Fig. 4.The detailed process that this absolute coordinate extracts comprises the steps:

S401, collection vehicle steering wheel first frame the second image information in default tracing area；

S402, in first frame the second image information, calculate the absolute coordinate of desired guiding trajectory starting point of steering wheel for vehicle；

S403, it is calculated the absolute coordinate of the desired guiding trajectory starting point in first frame the second image information processing according to Kalman filtering algorithm, obtains the absolute coordinate of next frame the second image information；

S404, when often receiving frame second image information, extract the absolute coordinate of desired guiding trajectory starting point in the second image information, obtain the absolute coordinate of desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the desired guiding trajectory starting point in all of second image information of output collection and the second image information.

Same, after extracting the second image information and absolute coordinate thereof, pass through track combination, it is possible to be easily obtained the second running orbit corresponding to steering wheel for vehicle.

The vehicle drive control method that the embodiment of the present invention provides carries out corresponding vehicle driving model switching by the interactive information of the first running orbit that driver's both hands are corresponding second running orbit corresponding with steering wheel for vehicle, referring to Fig. 5, this mode handover procedure specifically includes following steps:

S501, the first matching degree according to the first running orbit and the second running orbit, generate first driving model switching signal corresponding with the first matching degree；First driving model switching signal is used for controlling vehicle and is switched to pilot steering pattern by automatic driving mode；

S502, the second matching degree according to the first running orbit and the second running orbit, generate second driving model switching signal corresponding with the second matching degree；Second driving model switching signal is used for controlling vehicle and is switched to automatic driving mode by pilot steering pattern.

Wherein, control vehicle by the driving model switching signal that above-mentioned matching degree is corresponding and carry out corresponding driving model switching.

Additionally, under pilot steering pattern, the driving behavior of driver is divided into by analyzing the first running orbit and the second running orbit:

1, having turned to controlling behavior, now, the first running orbit and the second running orbit exist simultaneously and are all rendered as ellipse；

2, without having turned to controlling behavior, now, the first running orbit and the second running orbit exist simultaneously and are all approximately a bit；

3, without controlling behavior, now, it is absent from the first running orbit, there is the second running orbit.

Under automatic driving mode, by analyzing the first running orbit and the second running orbit, the driving behavior of driver is divided into:

1, entirely without controlling intention behavior, now, it is absent from the first running orbit；

2, there is control to be intended to, now, there is the first running orbit, but this first running orbit and the second running orbit occur simultaneously without obvious track；

3, being intended to control, the first running orbit and the second running orbit exist simultaneously.

By the further analysis of above-mentioned driver's driving behavior, understanding the intention of driver in real time accurately, this driver is intended to the corresponding bus that controls of cooperation to control switching and the maintenance of above-mentioned driving model.

The embodiment of the present invention additionally provides a kind of vehicle drive and controls system, and described system is used for performing above-mentioned vehicle drive control method, and referring to Fig. 6, described system includes:

First image information follows the tracks of acquisition module 11, and during for driver's both hands information being detected in default tracing area, real-time tracking gathers driver's both hands multiframe the first image information in default tracing area；

First analysis and processing module 22, multiframe the first image information gathered for the first image information is followed the tracks of acquisition module 11 is analyzed processing, and obtains the first running orbit of driver's both hands；

Second image information follows the tracks of acquisition module 33, for real-time tracking collection vehicle steering wheel multiframe the second image information in default tracing area；

Second analysis and processing module 44, multiframe the second image information gathered for the second image information is followed the tracks of acquisition module 33 is analyzed processing, and obtains the second running orbit of steering wheel for vehicle；

Pattern switching control module 55, controls the switching of vehicle implementation pattern for the switch mode processing, according to the first analysis and processing module 22, the first running orbit of obtaining and the second analysis and processing module 44 processes the matching degree of the second running orbit obtained corresponding；Wherein, pattern includes automatic driving mode and pilot steering pattern.

Further, in order to better realize the trajectory extraction to driver's both hands, referring to Fig. 7, above-mentioned first image information is followed the tracks of acquisition module 11 and is included the first image information collecting unit 111, fisrt feature information extraction unit 112 and location point computing unit 113, wherein

First image information collecting unit 111, for gathering driver's both hands first frame the first image information in default tracing area；

Fisrt feature information extraction unit 112, for when the first image information collecting unit 111 collects first frame the first image information, extracting the fisrt feature information in the first image information；Fisrt feature information includes the following information of driver's both hands: the first color characteristic information and the first body dynamics information；First body dynamics information at least includes the primary importance point of described driver's both hands；

Location point computing unit 113, for being calculated processing to the fisrt feature information that fisrt feature information extraction unit 112 is extracted according to the particle filter algorithm in multiple characteristics space, obtains the primary importance point of next frame the first image information；

Fisrt feature information extraction unit 112 is additionally operable to, and when often receiving frame first image information, extracts the fisrt feature information in the first image information；Location point computing unit 113 is additionally operable to, the fisrt feature information that fisrt feature information extraction unit 112 is extracted is calculated processing by the particle filter algorithm according to multiple characteristics space, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that the first image information is corresponding.

Further, consider that driver's both hands are as compliant motion target, in carrying out the process of image procossing being correlated with, being highly susceptible to the restriction of soft objectives self character and the situations such as loss occur to follow the tracks of, the vehicle drive that the embodiment of the present invention provides controls system and preferably adopts the tracking that the particle filter algorithm based on above-mentioned double characteristic information carries out follow-up driver's both hands to process.Then above-mentioned location point computing unit 113 includes target search particle and arranges subelement 1131, second feature information retrieval subelement 1132, color similarity computation subunit 1133 and location point computation subunit 1134, referring to Fig. 8, wherein,

Target search particle arranges subelement 1131, for arranging multiple target search particle meeting Gauss distribution in the preset range region of next frame the first image information；Wherein, the search graph picture that each target search particle is corresponding different respectively, search graph picture is the image of zones of different in next frame the first image information；

Second feature information retrieval subelement 1132, for extracting the second feature information of each search graph picture；Second feature information includes the following information of search graph picture: the second color characteristic information and the second body dynamics information；Second body dynamics information at least includes the second position point of search graph picture；

Color similarity computation subunit 1133, the the second color characteristic information extracted for the first color characteristic information that fisrt feature information extraction unit 112 is extracted and second feature information retrieval subelement 1132 carries out color similarity calculating, obtains the second color characteristic information the highest with the first color characteristic information similarity；

Location point computation subunit 1134, is defined as the primary importance point of driver's both hands in next frame the first image information for the second position point of search graph picture corresponding to the second color characteristic information of color similarity computation subunit 1133 being obtained.

Further, information in order to better realize driver's both hands and steering wheel for vehicle is mutual, referring to Fig. 9, the second image information in the vehicle drive control system that the embodiment of the present invention provides is followed the tracks of acquisition module 33 and is included the second image information collecting unit the 331, first absolute coordinate computing unit 332 and the second absolute coordinate computing unit 333, wherein

Second image information collecting unit 331, for the collection vehicle steering wheel first frame the second image information in default tracing area；

First absolute coordinate computing unit 332, for, in first frame the second image information that the second image information collecting unit 331 gathers, calculating the absolute coordinate of the desired guiding trajectory starting point of steering wheel for vehicle；

Second absolute coordinate computing unit 333, for being calculated processing to the absolute coordinate that the first absolute coordinate computing unit 332 calculates according to Kalman filtering algorithm, obtains the absolute coordinate of the desired guiding trajectory starting point of next frame the second image information；

First absolute coordinate computing unit 332 is additionally operable to, and when often receiving frame second image information, extracts the absolute coordinate of desired guiding trajectory starting point in the second image information；Second absolute coordinate computing unit 333 is additionally operable to, process according to the absolute coordinate that the first absolute coordinate computing unit 332 calculates, obtain the absolute coordinate of desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the desired guiding trajectory starting point in all of second image information of output collection and the second image information.

The vehicle drive that the embodiment of the present invention provides controls system and carries out corresponding vehicle driving model switching by pattern switching control module 55, referring to Figure 10, this pattern switching control module 55 includes the first switching signal and generates unit 551 and the second switching signal generation unit 552, wherein

First switching signal generates unit 551, process the first matching degree of the second running orbit obtained for the first running orbit obtained according to the first analysis and processing module 22 process and the second analysis and processing module 44, generate first driving model switching signal corresponding with the first matching degree；First driving model switching signal is used for controlling vehicle and is switched to pilot steering pattern by automatic driving mode；

Second switching signal generates unit 552, process the second matching degree of the second running orbit obtained for the first running orbit obtained according to the first analysis and processing module 22 process and the second analysis and processing module 44, generate second driving model switching signal corresponding with the second matching degree；Second driving model switching signal is used for controlling vehicle and is switched to automatic driving mode by pilot steering pattern.

The vehicle drive that the embodiment of the present invention provides controls system, first real-time tracking gathers driver's both hands and steering wheel for vehicle multiframe the first image information corresponding in default tracing area and multiframe the second image information, and is calculated above-mentioned multiframe the first image information and multiframe the second image information respectively processing the first running orbit obtaining correspondence and the second running orbit；Then the switching of vehicle implementation pattern is controlled according to the matching degree of the first running orbit and the second running orbit, with the control mode that employing torque sensor of the prior art carries out mode motor switching, security reliability is poor to be compared, it extracts the trace information of driver's hand and vehicle steering simultaneously, and can by the mutual corresponding driver intention of both trace informations for being applied on vehicle and this vehicle being controlled, the switch safety making driving model is higher, and achieve the flexible switching of automatic Pilot and pilot steering and harmonious mutual, also make practicality stronger.

In several embodiments provided herein, it should be understood that disclosed system, apparatus and method, it is possible to realize by another way.Device embodiment described above is merely schematic, such as, the division of described unit, it is only a kind of logic function to divide, actual can have other dividing mode when realizing, again such as, multiple unit or assembly can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some communication interfaces, device or unit, it is possible to be electrical, machinery or other form.

The described unit illustrated as separating component can be or may not be physically separate, and the parts shown as unit can be or may not be physical location, namely may be located at a place, or can also be distributed on multiple NE.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.

It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to be that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.

If described function is using the form realization of SFU software functional unit and as independent production marketing or use, it is possible to be stored in a computer read/write memory medium.Based on such understanding, part or the part of this technical scheme that prior art is contributed by technical scheme substantially in other words can embody with the form of software product, this computer software product is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-OnlyMemory), the various media that can store program code such as random access memory (RAM, RandomAccessMemory), magnetic disc or CD.

The above; being only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, any those familiar with the art is in the technical scope that the invention discloses; change can be readily occurred in or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.

Claims

1. a vehicle drive control method, it is characterised in that including:

When driver's both hands information being detected in default tracing area, real-time tracking gathers described driver's both hands multiframe the first image information in described default tracing area；

It is analyzed processing to described multiframe the first image information gathered, obtains the first running orbit of described driver's both hands；

Real-time tracking gathers described steering wheel for vehicle multiframe the second image information in described default tracing area；

It is analyzed processing to described multiframe the second image information gathered, obtains the second running orbit of described steering wheel for vehicle；

The switching of described vehicle implementation pattern is controlled according to the switch mode that described first running orbit is corresponding with the matching degree of described second running orbit；Wherein, described pattern includes automatic driving mode and pilot steering pattern.

2. vehicle drive control method according to claim 1, it is characterised in that described real-time tracking gathers described driver's both hands multiframe the first image information in described default tracing area and includes:

Gather the described driver's both hands first frame the first image information in described default tracing area；

When the described driver's both hands first frame the first image information in described default tracing area being detected, extract the fisrt feature information in described first image information；Described fisrt feature information includes the following information of described driver's both hands: the first color characteristic information and the first body dynamics information；Described first body dynamics information at least includes the primary importance point of described driver's both hands；

The fisrt feature information of described first frame the first image information is calculated processing by the particle filter algorithm according to multiple characteristics space, obtains the primary importance point of next frame the first image information；

When often receiving frame first image information, extract the described fisrt feature information in described first image information, and described fisrt feature information is calculated processing by the particle filter algorithm according to described multiple characteristics space, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that described first image information is corresponding.

3. vehicle drive control method according to claim 2, it is characterized in that, the fisrt feature information of described first frame the first image information is calculated processing by the described particle filter algorithm according to multiple characteristics space, and the primary importance point obtaining next frame the first image information includes:

Multiple target search particle meeting Gauss distribution is set in the preset range region of next frame the first image information；Wherein, the search graph picture that each described target search particle is corresponding different respectively, described search graph picture is the image of zones of different in described next frame the first image information；

Extract the second feature information of each described search graph picture；Described second feature information includes the following information of described search graph picture: the second color characteristic information and the second body dynamics information；Described second body dynamics information at least includes the second position point of described search graph picture；

Described first color characteristic information and described second color characteristic information are carried out color similarity calculating, obtains the second color characteristic information the highest with described first color characteristic information similarity；

The second position point of search graph picture corresponding for the described second color characteristic information obtained is defined as the primary importance point of driver's both hands in next frame the first image information.

4. vehicle drive control method according to claim 3, it is characterised in that described real-time tracking gathers described steering wheel for vehicle multiframe the second image information in described default tracing area and includes:

Gather the described steering wheel for vehicle first frame the second image information in described default tracing area；

In described first frame the second image information, calculate the absolute coordinate of the desired guiding trajectory starting point of described steering wheel for vehicle；

It is calculated processing to the absolute coordinate of the described desired guiding trajectory starting point in described first frame the second image information according to Kalman filtering algorithm, obtains the absolute coordinate of described next frame the second image information；

When often receiving frame second image information, extract the absolute coordinate of described desired guiding trajectory starting point in described second image information, obtain the absolute coordinate of described desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the described desired guiding trajectory starting point in all of second image information of output collection and described second image information.

5. vehicle drive control method according to claim 4, it is characterised in that the described switch mode corresponding with the matching degree of described second running orbit according to described first running orbit controls the switching of described vehicle implementation pattern and include:

The first matching degree according to described first running orbit and described second running orbit, generates first driving model switching signal corresponding with described first matching degree；Described first driving model switching signal is used for controlling described vehicle and is switched to pilot steering pattern by automatic driving mode；

The second matching degree according to described first running orbit and described second running orbit, generates second driving model switching signal corresponding with described second matching degree；Described second driving model switching signal is used for controlling described vehicle and is switched to automatic driving mode by pilot steering pattern.

6. a vehicle drive controls system, it is characterised in that including:

First image information follows the tracks of acquisition module, and during for driver's both hands information being detected in default tracing area, real-time tracking gathers described driver's both hands multiframe the first image information in described default tracing area；

First analysis and processing module, described multiframe the first image information for described first image information is followed the tracks of acquisition module collection is analyzed processing, and obtains the first running orbit of described driver's both hands；

Second image information follows the tracks of acquisition module, gathers described steering wheel for vehicle multiframe the second image information in described default tracing area for real-time tracking；

Second analysis and processing module, described multiframe the second image information for described second image information is followed the tracks of acquisition module collection is analyzed processing, and obtains the second running orbit of described steering wheel for vehicle；

Pattern switching control module, controls the switching of described vehicle implementation pattern for the switch mode processing described first running orbit obtained corresponding with the matching degree that described second analysis and processing module processes described second running orbit obtained according to described first analysis and processing module；Wherein, described pattern includes automatic driving mode and pilot steering pattern.

7. vehicle drive according to claim 6 controls system, it is characterised in that described first image information is followed the tracks of acquisition module and included:

First image information collecting unit, for gathering the described driver's both hands first frame the first image information in described default tracing area；

Fisrt feature information extraction unit, for when described first image information collecting unit collects described first frame the first image information, extracting the fisrt feature information in described first image information；Described fisrt feature information includes the following information of described driver's both hands: the first color characteristic information and the first body dynamics information；Described first body dynamics information at least includes the primary importance point of described driver's both hands；

Location point computing unit, for being calculated processing to the fisrt feature information that described fisrt feature information extraction unit is extracted according to the particle filter algorithm in multiple characteristics space, obtains the primary importance point of next frame the first image information；

Described fisrt feature information extraction unit is additionally operable to, and when often receiving frame first image information, extracts the described fisrt feature information in described first image information；Described location point computing unit is additionally operable to, the described fisrt feature information that described fisrt feature information extraction unit is extracted by the particle filter algorithm according to described multiple characteristics space is calculated processing, obtain the primary importance point of next frame the first image information, until when collecting last frame the first image information, exporting all of first image information of collection and the primary importance point that described first image information is corresponding.

8. vehicle drive according to claim 7 controls system, it is characterised in that described location point computing unit includes:

Target search particle arranges subelement, for arranging multiple target search particle meeting Gauss distribution in the preset range region of next frame the first image information；Wherein, the search graph picture that each described target search particle is corresponding different respectively, described search graph picture is the image of zones of different in described next frame the first image information；

Second feature information retrieval subelement, for extracting the second feature information of each described search graph picture；Described second feature information includes the following information of described search graph picture: the second color characteristic information and the second body dynamics information；Described second body dynamics information at least includes the second position point of described search graph picture；

Color similarity computation subunit, the described second color characteristic information extracted for the described first color characteristic information that described fisrt feature information extraction unit is extracted and described second feature information retrieval subelement carries out color similarity calculating, obtains the second color characteristic information the highest with described first color characteristic information similarity；

Location point computation subunit, is defined as the primary importance point of driver's both hands in next frame the first image information for the second position point of search graph picture corresponding to the described second color characteristic information that described color similarity computation subunit obtained.

9. vehicle drive according to claim 8 controls system, it is characterised in that described second image information is followed the tracks of acquisition module and included:

Second image information collecting unit, for gathering the described steering wheel for vehicle first frame the second image information in described default tracing area；

First absolute coordinate computing unit, for, in described first frame the second image information that described second image information collecting unit gathers, calculating the absolute coordinate of the desired guiding trajectory starting point of described steering wheel for vehicle；

Second absolute coordinate computing unit, for being calculated processing to the absolute coordinate that described first absolute coordinate computing unit calculates according to Kalman filtering algorithm, obtains the absolute coordinate of the described desired guiding trajectory starting point of described next frame the second image information；

Described first absolute coordinate computing unit is additionally operable to, and when often receiving frame second image information, extracts the absolute coordinate of described desired guiding trajectory starting point in described second image information；Described second absolute coordinate computing unit is additionally operable to, process according to the absolute coordinate that described first absolute coordinate computing unit calculates, obtain the absolute coordinate of described desired guiding trajectory starting point in next frame the second image information, until when collecting last frame the second image information, the absolute coordinate of the described desired guiding trajectory starting point in all of second image information of output collection and described second image information.

10. vehicle drive according to claim 9 controls system, it is characterised in that described pattern switching control module includes:

First switching signal generates unit, process the first matching degree of described second running orbit obtained for described first running orbit obtained according to described first analysis and processing module process and described second analysis and processing module, generate first driving model switching signal corresponding with described first matching degree；Described first driving model switching signal is used for controlling described vehicle and is switched to pilot steering pattern by automatic driving mode；

Second switching signal generates unit, process the second matching degree of described second running orbit obtained for described first running orbit obtained according to described first analysis and processing module process and described second analysis and processing module, generate second driving model switching signal corresponding with described second matching degree；Described second driving model switching signal is used for controlling described vehicle and is switched to automatic driving mode by pilot steering pattern.