WO2018210382A1

WO2018210382A1 - Method and device for a target-based prediction of dynamic objects

Info

Publication number: WO2018210382A1
Application number: PCT/DE2018/200001
Authority: WO
Inventors: Stefan Kerscher; Norbert Balbierer; Sebastian KRAUST; Nikolaus Müller
Original assignee: Continental Automotive Gmbh
Priority date: 2017-05-16
Filing date: 2018-01-19
Publication date: 2018-11-22
Also published as: DE112018000117A5; CN110168312B; DE102017208245A1; CN110168312A

Abstract

The invention relates to a method for estimating a future trajectory of an object, having the steps of: - generating target positions of the object, - determining the probability that the object is headed towards the respective target position, - selecting the target positions with the highest degree of probability and defining same as the target of the object, and - determining the trajectory of the object from the current position thereof relative to the target.

Description

Method and device for the goal-based prediction of dynamic objects

The invention relates to a method and a device for the goal-based prediction of dynamic object. Based on the prediction of a future goal, a

Trajectory of the object is determined and this trajectory used to control a vehicle.

State of the art

From DE 10 2008 049 824 B4 the prediction of

Movement of persons known to carry a mobile phone. Here, the recognition of the

prediction object realized by receiving messages from the mobile phone. From the history of

received messages, a trajectory of the object is determined.

EP 2 562 060 B1 discloses the prediction of dynamic objects on highways, where highways represent a very well-structured environment. Furthermore, future maneuvers of road users are estimated from the context and measured movements with internally existing ones

Motion models compared. The two won

Statements are compared for agreement and extra movement is extrapolated from it.

EP 2 382 609 B1 shows a method for predicting movements by variation of movement quantities by a pre-trained model, considering all physically possible movements and subsequently restricted by house walls, road courses, etc. Task and solution

Object of the present invention is the

Predictability of the trajectory of objects to improve. The object is achieved by a method according to the independent claim and a system according to the

In addition claim. Further advantageous embodiments of the invention are the subject of the dependent claims.

The inventive method for estimating a

future trajectory of an object has the following

Steps on, generating target positions of the object, determining a probability that the object is heading to the respective target position, selecting the

Target position with the highest probability and

Set as the object's destination and determine a

Trajectory of the object from its current position to the target.

Preferably, the probability by means of a

Bayesian network, or neural network, or support vector machine, or cluster analysis.

Advantageously, the probabilities for driving the potential target positions through the object can be calculated and made comparable by means of these methods. In a further embodiment of the invention, the trajectory of the object by air line or A * - algorithm or potential field method or sampling

Methods or state lattice or rapid exploring random trees are determined. Advantageously, the trajectory can be used as a function of the available information with these methods calculate and, if appropriate, adjust continuously.

Advantageously, a point of interest is a place which, by virtue of particular properties, has a high attraction to objects of a particular class.

For example, it may be a pedestrian crossing, a crosswalk or a pedestrian bus stop.

Further preferably, the probability may be adjusted by classifying the target position if the target position is a point of interest or within the motion environment of the object or a path to

Target position is locked or traffic laws or

Traffic signs prohibit reaching the target position.

In an advantageous manner, the determined

Trajectory will be further improved because of

Obstacles or traffic signs are adapted to impossible trajectories, for example, the physical movement possibilities of a pedestrian or the

Acceleration possibility of a vehicle are considered. Furthermore, traffic rules can be taken into account, for example, is at pedestrian crossings with

Pedestrians, but not on highways.

Preferably, an uncertainty factor of the trajectory can be determined from the probability. In an advantageous manner, such an evaluation of

Trajectory determined to detect any deviations of the object from the trajectory early on.

Further preferably, the uncertainty factor may be further filtered by a Kalman Filter or Unscented Kalman Filter or

Extended caiman filters or particle filters are determined. Advantageously, these filters allow one

continuously calculating the deviation of the trajectory from the original trajectory, so that if the deviation is too great, a correction can be made. In a further embodiment of the invention, the

Target positions come from a navigation system or be known as the point of interest of the object.

Advantageously, such known information from maps, such as contour lines, or others

Sources are used for the prediction of the target positions.

Preferably, the target positions by

safety-relevant aspects or usual

motion-relevant aspects are generated. Advantageously, for example, physiological limits of a human or pedestrian crossings can be taken into account in the selection of the potential target positions.

More preferably, the probability can be continuously determined and the trajectory adapted to the probability.

Advantageously, therefore, is a continuous

Forecasting the trajectory based on the most probable target position possible. According to the invention, a system in a vehicle having an environment detection system is set up to carry out a method according to one of the preceding preferred embodiments.

The surroundings detection system may preferably comprise at least one radar sensor or a camera or a LIDAR. figure description

In a first step, possible target positions of the objects are generated. In Figures 1 to 5 are for

Examples shown. Figure 1 shows an example of generating a

Target position of a pedestrian. The vehicle 1 moves within its driving tube 3. Outside the road 11, a pedestrian is recognized as an object 7. The pedestrian 7 could traverse the driving tube 3 of the vehicle 1.

For a possible target position 9 of the pedestrian 7 is generated such that a vertical axis 15 is 13 formed on the center line 5 of the driving tube 3 and the target position 9 of the pedestrian 7 from the center line 5 has the same distance as the pedestrian 7 of the

Center line 5 of the driving tube 3.

FIG. 2 shows another example of the generation of a target position of a pedestrian. The vehicle 1 moves within its driving tube 3. A pedestrian is recognized as an object 7 within the driving tube 3. The pedestrian 7 could leave the driving tube 3 of the vehicle 1 in two directions. Two possible target position 9, 11 of the pedestrian 7 are generated such that a

vertical axis 13 is formed on the center line 5 of the driving tube 3 and the target positions 9, 10 of the pedestrian 7 are outside the driving tube 3.

FIG. 3 shows an example of generating a

Target position of a standing pedestrian. The vehicle 1 moves within its driving tube 3. Outside the road 11, a pedestrian is recognized as an object 17. The pedestrian 7 could stand on the side of the road and not move. For this case, a possible target position 19 of the pedestrian 17 is generated at the location of the pedestrian 19. FIG. 4 shows another example of the generation of target positions of a pedestrian. The vehicle 1 moves within its driving tube 3. Outside the road 11, a pedestrian is recognized as an object 7. The pedestrian 7 could go along the street 11. It will therefore

possible target positions 10, 10 a generated on the vehicle side of the pedestrian 7. Further still possible

Target positions 9, 9a on the opposite

Vehicle side generated because these target positions could be interesting for more pedestrians. The

Target positions 9, 9 a could pedestrian 7 through a

Passing through the driving tube 3 reach and so on

Combination with the example of Figure 1 can be achieved.

FIG. 5 shows another example of the generation of a target position of a pedestrian. The vehicle 1 moves within its driving tube 3. Outside the road 11, a pedestrian is recognized as an object 7. The pedestrian 7 could move away from the street 11. The target position 21 is therefore by a vertical axis 15 on the

Center line 5 of the driving tube 3 is formed and the distance of the target position 21 from the center line 5 is twice as large as the distance of the pedestrian 7 from the center line. 5

In a second step, probabilities for driving the potential target positions are determined.

FIG. 6 shows an example of classifying the

Target positions according to the respective probability for driving through the object. Pictured is a

Bayesian network 40 and evaluated an assumed destination point 41 by classification in different categories.

Inputs for the Bayesian network are, for example, a pedestrian and its possible target positions 41 (target point, TP). For each possible target position 41 is the WWaahhrrsscchheeillllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll

EEss wwiirrdd aallssoo overruns, oobb ddiiee ZZiieellppoossiittiioonn 4411

bbeeiissppiieellsswweeiissee iinn eeiinneemm ffuußßggäännggeerrttyyppiisscchheenn BBeerreeiicchh lliieeggtt 4422 55 ((IISS oonn ppeeddeesstt aarreeaa)) ,, tthhee ooddeerr oobb FFuußßggäännggeerr iinn tthhee RRiicchhttuunngg ZZiieellppoossiittiioonn sscchhaauutt 4433 ((hheeaaddiinngg ttoo TTPP)) ,, tthhee ooddeerr oobb

FFiussgggangnggererr ssiicchh iinn RRiicchhttuunngg ddeerr ZZiieellppoossiittiioonn ddrreehtt 4444

((yyaawwrraattee ttoo TTPP)) ,, ooddeerr oobb ddeerr FFuussßgggnnggeerr sseeiinnee

GGeesscchhwwiinnddiiggkkeeiittsskkoommppoonneennttee iinn tthhee RRiicchhttuunngg ZZiieellppoossiittiioonn hhaatt 1100 4455 ((vveelloocciittyy ttoo TTPP)) ,, tthhee ooddeerr oobb FFuußßggäännggeerr iinn tthhee RRiicchhttuunngg ZZiieellppoossiittiioonn bbeesscchhlleeuunniiggtt 4466 ((aacccceelleerraattiioonn ttoo TTPP)) ,, ooddeerr oobb ooff tthhee FFuußßggäännggeerr AAMM EErrrreeiicchheenn ZZiieellppoossiittiioonn dduurrcchh eeiinnee VVeerrkkeehhrrssrreeggeell ggeehhiinnddeerrtt iisstt 4477 ((bblloocckkeedd bbyy llaaww)) ,, ooddeerr oobb ddeerr FFgussggangangergeerr aamm EErrrreeiicchheenn dZerieppiossiittiioonn dduurrcchh eeiinn

1155 HHiinnddererrnniiss ggeehhiinnddeerrtt iisstt 4488 ((bblloocckkeedd bbyy oobbjjeecctt)) ,, ooddeerr oobb ddiiee ZZiieellppoossiittiioonn iinn ddeerr NNäähhee eeiinneess fdürenn FFgussggännggeerr

In the middle of the year, it was published in the German edition of the German Federal Foreign Office in 1999. (oiitdeerr oobb ddeerr FFgussggangnggeerr the ZZiieellppoossiittiioonn hhaartn 5511 ((iss ssttaayy ppooiinntt)), eerr allall ssttaattiisscchh iisstt ..

2200 I nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn.

In the eeiinneemm vviieerrtteenn SScchhrriitttt eerrffoollggtt eeiinnee

UUnnssiicchheerrhheeiittssaabbsscchhäättzzuunngg tthhee bbeessttiimmmmtteenn TTrraajjeekkttoorriiee .. TThhee iinn ddeenn FFiigguurreenn 77 bbiiss 1100 ddaarrggeesstteellllttee UUnnssiicchheerrhheeiittssaabbsscchhäättzzuunngg 2255 nniimmmmtt bbeeiissppiieellsswweeiissee AAnn ,, ddaassss eeiinn FFuußßggäännggeerr aallss OObbjjeekktt ssiicchh sstteettss aauuff ddeemm kküürrzzeesstteenn WWeegg zzuu sseeiinneemm ZZiieell bbeewweeggeenn wwiirrdd ..

FFiigguurr 77 zzeeiiggtt eeiinnee bbeessttiimmmmttee TTrraajjeekkttoorriiee bbeesstteehhtt ffrroomm mmeehhrreerreenn TTrraajj eekkttoorriieennppuunnkktteenn 6611 ,, 6633 ,, 6655 zzwwiisscchheenn SSttaannddoorrtt XXOO tthhee OObbjjeekktteess aanndd ddeemm eerrmmiitttteelltteenn ZZiieell TTOO ,, ffoorr wwoobbeeii

3300 jjeeddeenn TTrraajj eekkttoorriieennppuunnkktt 6611 ,, 6633 ,, 6655 eeiinnee KKoovvaarriiaannzzmmaattrriixx

bbeerreecchhnneett wwiirrdd .. WWiirrdd ddaass OObbjjeekktt XXOO fiercely irritated MMaall

ddeetteekkttiieerrtt ,, reffilterednessnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn Way to the destination TO is planned. Also the covariance matrix of the estimated uncertainty becomes

All

TTrraajj eekkttoorriieennppuunnkkttee 6611 ,, 6633 ,, 6655 ssiinndd mmiitt ddeenn

Uunnssiicchheerrhheeiitteenn bbeelleeggtt ,, wwaass iinn ddeerr FFiigguurr 77 dduurrcchh KKrreeiisseenn uumm 55 ddiiee TTrraajj eekkttoorriieennppuunnkkttee 6611 ,, 6633 ,, 6655 vveerrddeeuuttlliicchhtt wwiirrdd ..

FFiigguurr 88 zzeeiigg tthhee wweeiitteerree BBeessttiimmmmuunngg tthhee TTrraajjeekkttoorriiee nnaacchh eeiinneemm ZZeeiittsscchhrriitttt kk ,, nnaacchhddeemm eeiinnee nneeuuee MMeessssuunngg tthhee OObbjjeekktteess AAnn tthhee PPoossiittiioonn XXII ggeemmaacchhtt wwuurrddee .. DDiieessee MMeessssuunngg iisstt iinn tthhee FFiigguurr 88 uunntteerrhhaallbb tthhee PPffaaddeess tthhee TTrraajjeekkttoorriiee ggeezzeeiiggtt .. ddaa

In the year 1100 the AAkkttuuaalliissiieerrunnggssrraattee nniicchhttnnoottwweennddiiggeerr WWeeiissee mmiitt ddeerr

zzeeiittlliicchheenn DDiissttaannzz zzwwiisscchheenn ddtrdndrajj eekkttoorriieennppuunnkktteenn 6611 ,, 6633 ,, 6655 ddeess kuurrzezzeesstteenn PPffaaddss üübbeerreeiinnssttiimmmenn mmuussss ,, wwiirrdd eeiinnee

PPoossiittiioonn XXlliinntt iinn ddeerr vvoorrmmaallss bbeerreecchhnneetteenn TTrraajjeekkttoorriiee eerrmmiittteltt ,, which is zzeeiittlliicchh zzuurr PPoossiittiioonn XXII ppaassstt .. DDiieessee

1155 iinntteerrppoolliieerrttee PPoossiittiioonn XXlliinntt wwiirrdd aallss KKaaiimmaann ZZuussttaanndd

ggeesseettzztt .. IInn ddeerr Figigguurr 88 bbeeffiinnddeett ssiicchh ddiiee PPoossiittiioonn XXlliinntt zzwwiisscchheenn XXOO and ddeemm nnääcchhstteenn TTrraajj eekkttoorriieennppuunnkktt 6611 ..

AAnnsscchhlliieeßßenndd wwiirrdd a UUppddaattee - SScchhrriitttt ddeess KKaaiimmaann FFiilltteerrss aauussggeefführhrtt uunndd ssoommiitt ddiiee PPoossiittiioonn XXlluuppddaattee bbeerreecchhnneett ,,

2200 wweellcchhee iinn FFiigguurr 99 ddaarrggeesstteelllltt iisstt .. TThhee PPoossiittiioonn XXlluuppddaattee bbeeffiinnddeett ssiicchh zzwwiisscchheenn XXlliinntt aanndd XXII .. TThhee AAbbssttaanndd ooff XXlliinntt aanndd XXlluuppddaattee iisstt dduurrcchh TThhee SSyysstteemmrraauusscchheenn bbeeddiinnggtt QQ .. TThhee SSyysstteemmrraauusscchheenn QQ kkaannnn dduurrcchh tthhee ffoollggeennddee MMaattrriixx bbeerreecchhnneett wweerrddeenn .. TThhee VVaarriiaabblleenn ssiinndd wwiiee ffoollggtt ddeeffiinniieerrtt :: KK does not ddeerr

In 2255, it is at the same time that it is used for the purpose of preserving, and for the sake of, its purpose

Room.

By changing Q, the changing uncertainty of the prediction can be modeled. The worse the

Prediction matches the actual measured states, the more uncertain the prediction and the system noise Q increases. If an object follows the prediction well, the predicated states differ little from the reality and the system noise Q remains small.

FIG. 10 shows an update of the trajectory of the object to the target formed from Q. From the new

estimated position Xlupdate becomes a trajectory to

Target TO calculated. For each trajectory point 61, 63, 65, the prediction step of the Kalman filter is performed for the covariance matrices. The resulting positions 62, 64, 66 of the covariance matrix are entered in FIG. 10 with error ellipses.

Claims

claims

A method for estimating a future trajectory of an object comprising the steps of:

Generating target positions of the object,

Determining a probability that the object will drive the respective target position,

- Select the target position with the highest

Probability and setting as the object's goal, and

Determining a trajectory of the object of

his current position to the goal.

2. The method according to claim 1, wherein the

Probability by means of a

- Bayesian network, or

- Neural network, or

- Support Vector Machine, or

- Cluster analysis

is determined.

3. The method according to any one of the preceding claims, wherein the trajectory of the object by

- air line, or

- A * algorithm, or

- Potential field method, or

- Sampling methods, or

- State Lattice, or

- Rapid exploring random trees

is determined.

4. The method according to one of the preceding claims, wherein the probability is further adjusted by classifying the target position when the

target position

- is a point of interest, or - lies within the movement environment of the object, or

- a path to the target position is blocked, or

- Traffic laws or traffic signs

Prohibit reaching the target position.

5. The method according to any one of the preceding claims, wherein an uncertainty factor of the trajectory from the

Probability is determined.

6. The method according to claim 5, wherein the

Uncertainty factor further by a

- Kaiman filter, or

- Unscented Cayman Filter, or

- Extended Cayman Filter, or

- Particle filter

is determined.

7. The method according to any one of the preceding claims, wherein the destination positions originate from a navigation system or are known as the point of interest of the object.

8. The method according to any one of the preceding claims, wherein the target positions by

- security aspects, or

- usual movement-relevant aspects

to be generated.

9. The method according to one of the preceding claims, wherein the probability is continuously determined and the trajectory is adapted to the probability.

10. System in a vehicle with a

Environment detection system, which is adapted to perform a method according to one of the preceding claims.