EVALUATION APPARATUS FOR EVALUATING A TRAJECTORY HYPOTHESIS FOR A VEHICLE

Abstract

An evaluation apparatus for evaluating a predefined trajectory hypothesis for a vehicle includes a calculation unit, a detection unit and an assigning unit. The calculation unit is configured to calculate at least one necessary driving parameter for the vehicle to follow the predefined trajectory hypothesis. The detection unit is configured to detect a current driving parameter of the vehicle. The assigning unit is configured to assign a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

Claims

1. An evaluation apparatus for evaluating a predefined trajectory hypothesis for a vehicle, comprising: a calculation unit to calculate at least one necessary driving parameter for the vehicle to follow the predefined trajectory hypothesis; a detection unit to detect a current driving parameter of the vehicle; and an assigning unit to assign a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

2. The evaluation apparatus of claim 1, wherein the calculation unit is configured to calculate at least a second necessary driving parameter for the vehicle to follow a predefined second predefined trajectory hypothesis, and wherein the assigning unit is configured to assign a second probability value for the vehicle to follow the second predefined trajectory hypothesis using the current driving parameter and the second necessary driving parameter.

3. The evaluation apparatus of claim 2, wherein the assigning unit is configured to assign a higher probability value for the vehicle to follow the predefined trajectory hypothesis than for the vehicle to follow the second predefined trajectory hypothesis when the current driving parameter resembles the necessary driving parameter more than the second necessary driving parameter, and/or wherein the assigning unit is configured to assign a higher second probability value for the vehicle to follow the second predefined trajectory hypothesis than for the vehicle to follow the predefined trajectory hypothesis when the current driving parameter resembles the second necessary driving parameter more than the necessary driving parameter.

4. The evaluation apparatus of claim 1, wherein the calculation unit is configured to calculate the necessary driving parameter as a necessary steering wheel angle and/or a necessary acceleration for the vehicle to follow the predefined trajectory hypothesis and/or the detection unit is configured to detect the current driving parameter as an actuated steering wheel angle and/or an actuated acceleration of the vehicle.

5. The evaluation apparatus of claim 1, further comprising: a determining unit to determine at least the one predefined trajectory hypothesis for the vehicle using at least a sensor information of the vehicle, a driver monitoring information and/or an observed environmental information.

6. The evaluation apparatus of claim 1, further comprising: a collision calculation unit to calculate a collision possibility value for a collision of the vehicle with an object or further vehicle using the probability value.

7. The evaluation apparatus of claim 6, wherein the collision calculation unit is configured to calculate a collision avoidance maneuver for adapting the predefined trajectory hypothesis using the collision possibility value.

8. The evaluation apparatus of claim 1, further comprising: a selection unit to select the predefined trajectory hypothesis as a predicted trajectory for the vehicle to follow when the probability value reaches or exceeds a defined probability value or comprises the highest probability value out of several assigned probability values for several different predefined trajectory hypotheses.

9. The evaluation apparatus of claim 1, wherein the assigning unit is configured to assign the probability value as a function of the deviation of the necessary driving parameter and the current driving parameter, and/or at least one previous driving parameter of the vehicle actuated prior to the current driving parameter, and/or at least one previous necessary driving parameter of the vehicle calculated prior to the necessary driving parameter.

10. A vehicle, comprising: an evaluation apparatus for evaluating a predefined trajectory hypothesis for the vehicle, including: a calculation unit to calculate at least one necessary driving parameter for the vehicle to follow the predefined trajectory hypothesis; a detection unit to detect a current driving parameter of the vehicle; and an assigning unit to assign a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

11. A method of controlling an evaluation apparatus for evaluating a predefined trajectory hypothesis for a vehicle, the method comprising: calculating, via a calculation unit of the evaluation apparatus, at least one necessary driving parameter for the vehicle to follow the predefined trajectory hypothesis; detecting, via a detection unit of the evaluation apparatus, a current driving parameter of the vehicle; and assigning, via an assigning unit of the evaluation apparatus, a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

12. A non-transitory computer readable medium having a computer program, which is executable by a processor, comprising: a program code arrangement having program code for controlling an evaluation apparatus according to one of the preceding claims, by performing the following: calculating at least one necessary driving parameter for the vehicle to follow a predefined trajectory hypothesis; detecting a current driving parameter of the vehicle; and assigning a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

13. The computer readable medium of claim 12, wherein at least a second necessary driving parameter is calculated for the vehicle to follow a predefined second predefined trajectory hypothesis, and wherein the assigning includes assigning a second probability value for the vehicle to follow the second predefined trajectory hypothesis using the current driving parameter and the second necessary driving parameter.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0029] FIG. 1 shows a schematic illustration of a vehicle comprising an evaluation apparatus according to an embodiment of the present invention for evaluating a predefined trajectory hypothesis.

[0030] FIG. 2 shows a flowchart of a method of controlling an evaluation apparatus according to an embodiment of the present invention.

DETAILED DESCRIPTION

[0031] FIG. 1 shows a schematic illustration of a vehicle 100 comprising an evaluation apparatus 105 according to an embodiment of the present invention for evaluating a predefined trajectory hypothesis A. The vehicle 100 is a utility vehicle or commercial vehicle and/or configured as an autonomous vehicle 100 or a semi-autonomous vehicle 100. According to this embodiment, the vehicle 100 comprises at least one vehicle sensor 110.

[0032] The evaluation apparatus 105 for evaluating the predefined trajectory hypothesis A comprises a calculation unit 115, a detection unit 120 and an assigning unit 125. The calculation unit 115 is configured to calculate at least one necessary driving parameter ϕ.sub.A for the vehicle 100 to follow the predefined trajectory hypothesis A. The detection unit 120 is configured to detect a current driving parameter ϕ.sub.ego of the vehicle 100. The assigning unit 125 is configured to assign a probability value P(A) for the vehicle 100 to follow the predefined trajectory hypothesis A using the current driving parameter ϕ.sub.ego and the necessary driving parameter ϕ.sub.A.

[0033] According to this embodiment, the evaluation apparatus 105 is configured to receive at least the predefined trajectory hypothesis A or several different predefined trajectory hypotheses A, B from a determining unit 130 that is configured to determine at least the predefined trajectory hypothesis A for the vehicle 100 using for example at least a sensor information 135 of the vehicle 100, a driver monitoring information and/or an observed environmental information from the vehicle sensor 110. According to an embodiment, the sensor information 135 represents a yaw rate of the vehicle 100 and/or a vehicle velocity of the vehicle 100. For example, boundaries of the current vehicle lane serve as a source to predict the predefined trajectory hypothesis A and/or a second predefined trajectory hypothesis B. According to this embodiment, the determining unit 130 is part of the evaluation apparatus 105.

[0034] The predefined trajectory hypothesis A is a possible trajectory to be followed by the vehicle 100 out of a set of several possible trajectories. The necessary driving parameter ϕ.sub.A is a subsequent driving parameter which necessarily needs to be actuated in order to control the vehicle 100 to follow the predefined trajectory hypothesis A. The necessary driving parameter ϕ.sub.A represents for example an actuation value or control value of a vehicle setting unit of the vehicle 100. The current driving parameter ϕ.sub.ego represents an actuation value or control value which is actually actuated. The current driving parameter ϕ.sub.ego represents an actuation value or control value of the same vehicle setting unit of the vehicle 100, so that the current driving parameter ϕ.sub.ego and the necessary driving parameter ϕ.sub.A are comparable driving parameters. According to an embodiment, the assigning unit 125 is configured to assign the probability value P(A) for the vehicle 100 to follow the predefined trajectory hypothesis A using a result of a comparison between the current driving parameter ϕ.sub.ego and the necessary driving parameter ϕ.sub.A.

[0035] According to this embodiment, the calculation unit 115 is configured to calculate at least a second necessary driving parameter ϕ.sub.B for the vehicle 100 to follow a second predefined trajectory hypothesis B; wherein the assigning unit 125 is configured to assign a second probability value P(B) for the vehicle 100 to follow the second predefined trajectory hypothesis B using the current driving parameter ϕ.sub.ego and the second necessary driving parameter ϕ.sub.B.

[0036] According to this embodiment, the assigning unit 125 is configured to assign a higher probability value P(A) for the vehicle 100 to follow the predefined trajectory hypothesis A than for the vehicle 100 to follow the second predefined trajectory hypothesis B if the current driving parameter ϕ.sub.ego resembles the necessary driving parameter ϕ.sub.A more than the second necessary driving parameter ϕ.sub.B and/or wherein the assigning unit 125 is configured to assign a higher second probability value P(B) for the vehicle 100 to follow the second predefined trajectory hypothesis B than for the vehicle 100 to follow the predefined trajectory hypothesis A if the current driving parameter ϕ.sub.ego resembles the second necessary driving parameter ϕ.sub.B more than the necessary driving parameter ϕ.sub.A. According to this embodiment, the higher probability value P(A) for the vehicle 100 to follow the predefined trajectory hypothesis A is assigned if the difference between the current driving parameter ϕ.sub.ego and the necessary driving parameter ϕ.sub.A is lower than the difference between the current driving parameter ϕ.sub.ego and the second necessary driving parameter ϕ.sub.B. According to another embodiment, the higher second probability value P(B) for the vehicle 100 to follow the second predefined trajectory hypothesis B is assigned if the difference between the current driving parameter ϕ.sub.ego and the second necessary driving parameter ϕ.sub.B is lower than the difference between the current driving parameter ϕ.sub.ego and the necessary driving parameter ϕ.sub.A.

[0037] According to this embodiment, the calculation unit 115 is configured to calculate the necessary driving parameter ϕ.sub.A as a necessary steering wheel angle and/or a necessary acceleration for the vehicle 100 to follow the predefined trajectory hypothesis A and/or the detection unit 120 is configured to detect the current driving parameter ϕ.sub.ego as an actuated steering wheel angle and/or an actuated acceleration of the vehicle 100.

[0038] According to this embodiment, the evaluation apparatus 100 furthermore comprises a collision calculation unit 140, wherein the collision calculation unit 140 is configured to calculate a collision possibility value CP(A) for a collision of the vehicle 100 with an object or further vehicle using the probability value P(A). According to an embodiment, the collision calculation unit 140 is configured to calculate the collision possibility value CP(A) for the collision of the vehicle 100 with the object or further vehicle or further vehicle on the predefined trajectory hypothesis A using the probability value P(A). According to this embodiment, the collision calculation unit 140 is configured to calculate a second collision possibility value for a collision of the vehicle 100 with an object or further vehicle using the probability value P(B). According to an embodiment, the collision calculation unit 140 is configured to calculate the second collision possibility value for the collision of the vehicle 100 with the object or further vehicle on the second predefined trajectory hypothesis B using the probability value P(B).

[0039] According to this embodiment, the collision calculation unit 140 is configured to calculate a collision avoidance maneuver for adapting the predefined trajectory hypothesis A using the collision possibility value CP(A). According to an embodiment, the collision calculation unit 140 is furthermore configured to calculate a second collision avoidance maneuver for adapting the second predefined trajectory hypothesis B using the second collision possibility value.

[0040] According to an embodiment, the evaluation apparatus 100 furthermore comprises a selection unit, wherein the selection unit is configured to select the predefined trajectory hypothesis A as a predicted trajectory for the vehicle 100 to follow if the probability value P(A) reaches or exceeds a defined probability value or comprises the highest probability value out of several assigned probability values for several different predefined trajectory hypotheses A, B. According to an embodiment, the selection unit is configured to select the second predefined trajectory hypothesis B as the predicted trajectory for the vehicle 100 to follow if the second probability value P(B) reaches or exceeds a defined probability value or comprises the highest probability value out of several assigned probability values for several different predefined trajectory hypotheses A, B. According to an embodiment, the selection unit is configured to provide a control signal for a control unit of the vehicle 100, wherein the control signal is configured to control the vehicle 100 in order to follow the predicted trajectory. According to an embodiment, the evaluation apparatus 100 comprises the control unit that is configured to control the vehicle 100 in order to follow the predicted trajectory using the control signal.

[0041] According to an embodiment, the assigning unit 125 is configured to assign the probability value P(A) as a function of the deviation of the necessary driving parameter ϕ.sub.A and the current driving parameter ϕ.sub.ego and/or at least one previous driving parameter of the vehicle 100 actuated prior to the current driving parameter ϕ.sub.ego and/or at least one previous necessary driving parameter of the vehicle 100 calculated prior to the necessary driving parameter ϕ.sub.A. Examples for assigning the probability value P(A) and the second probability value P(B) are shown above:

P(A)=f(|ϕ.sub.ego−ϕ.sub.A|)

P(B)=f(|ϕ.sub.ego−ϕ.sub.B|)

[0042] In other words, FIG. 1 shows an architecture of an evaluation apparatus 105, which enables the usage of control values to estimate the probability of trajectory hypotheses A, B which may also be referred to as “ego motion trajectories” or “ego vehicle trajectories”.

[0043] According to an embodiment, the evaluation apparatus 105 is integrated or implemented into a driver assistant system of the vehicle 100, such as ACC (Adaptive Cruise Control), AEBS (Advanced Emergency Braking System) or PAEBS. For such driver assistant systems, the driver of the vehicle 100 partially or mainly defines the future state of the ego vehicle 100. To make decisions, those systems consider one or more trajectory hypotheses A, B of the predicted vehicle state for a certain prediction time period.

[0044] According to an embodiment, the hypothesis A, B of the future vehicle state for the predicted time period (ego vehicle trajectory) is estimated using different sources, such as sensor information 135 within the ego vehicle 100, driver monitoring information and/or observed environmental information. For example, if available, the boundaries of the current vehicle lane, is a source to predict the future vehicle trajectory. Another source is the current state of the vehicle dynamics, such as yaw rate and vehicle velocity. To make a decision based on the ego vehicle trajectory for each of the hypotheses A, B, a certain probability of occurrence is assigned/estimated by the assigning unit 125. According to an embodiment, based on the probability for example the trajectory hypothesis A, B with the highest probability is selected or the probability for each hypotheses A, B is considered directly when making decisions.

[0045] In the context of automated driving, control algorithms are used, to calculate the necessary actuation values, e. g. steering wheel angle, acceleration demand, . . . , based on a predicted trajectory.

[0046] Here, a basic idea is to use the actuation value of an autonomous vehicle controller in order to estimate the probability if the driver will follow a certain trajectory hypothesis A, B. In order to estimate the probability of occurrence, this functionality is applied to different ego vehicle trajectories A, B. For each of the trajectories A, B . . . the actuation values are calculated, that a control algorithm of an automated vehicle 100 would choose in order to follow the specific trajectory (ϕ.sub.A, ϕ.sub.B, . . . ). According to an embodiment, afterward, the estimated actuation value is compared to the actual actuation value.

[0047] According to an embodiment, the probability of occurrence of each of the ego vehicle trajectories A, B is calculated as a function of the deviation of the estimated actuation value and the actual actuation value and their values from the previous time step.

[0048] According to an embodiment, the probability of occurrence is then used in order to estimate the collision probability for the specific ego vehicle trajectory A, B, which afterwards can be used for e. g. collision avoidance maneuvers.

[0049] FIG. 2 shows a flowchart of a method 200 of controlling an evaluation apparatus according to an embodiment of the present invention. The method 200 is executable by the evaluation apparatus as described with reference to FIG. 1 or a similar evaluation apparatus.

[0050] The method 200 of controlling comprises a step 210 of calculating at least one necessary driving parameter for the vehicle to follow a predefined trajectory hypothesis. The method 200 of controlling furthermore comprises a step 220 of detecting a current driving parameter of the vehicle. Furthermore, the method 200 of controlling comprises a step 230 of assigning a probability value for the vehicle to follow the predefined trajectory hypothesis using the current driving parameter and the necessary driving parameter.

THE REFERENCE NUMERAL LIST IS AS FOLLOWS

[0051] ϕ.sub.A necessary driving parameter [0052] ϕ.sub.B second necessary driving parameter [0053] ϕ.sub.ego current driving parameter [0054] A predefined trajectory thesis [0055] B second predefined trajectory thesis [0056] CP(A) collision possibility value [0057] P(A) probability value [0058] P(B) second probability value [0059] 100 vehicle [0060] 105 evaluation apparatus [0061] 110 vehicle sensor [0062] 115 calculation unit [0063] 120 detection unit [0064] 125 assigning unit [0065] 130 determining unit [0066] 135 sensor information [0067] 140 collision calculation unit [0068] 200 method of controlling an evaluation apparatus for a vehicle [0069] 210 step of calculating [0070] 220 step of detecting [0071] 230 step of assigning