Computer-implemented method and system for planning the behavior of a vehicle

1 . A computer-implemented method of an automated driving system for controlling a vehicle, the method comprising the following steps:

(i) obtaining, by a processor system of the vehicle that includes at least one processor, a plurality of real-time sensor signals from one or more sensors of the vehicle sensing surroundings of the vehicle, thereby generating aggregated situation-specific information; and

(ii) generating, by the processor system, an environmental model of a current driving situation of the vehicle based on the aggregated situation-specific information, wherein the environmental model comprises a representation of physical surroundings relevant to the vehicle's operational context;

(iii) determining, by the processor system and based on the environmental model and a predefined set of one or more mandatory compliance rules, one or more boundary conditions defining mandatory constraints on permissible behaviors of the vehicle in the current driving situation of the vehicle;

(iv) based on the determined boundary conditions, eliminating, by the processor system, one or more of a plurality of possible behaviors from consideration in a subsequent behavior selection process;

(v) subsequent to the elimination, performing, by the processor system, the behavior selection process by applying a decision-making process structure comprising a predefined set of prioritization rules to remaining behaviors of the plurality of possible behaviors which have not been eliminated by the eliminating step, wherein:

the decision-making process structure comprises a hierarchical sequence of rule evaluations characterized by that the application of the decision-making process structure evaluates a conformity of the remaining behaviors to the predefined set of prioritization rules in an order that is defined by the hierarchical sequence according to relative priorities of the prioritization rules to one another, resulting in, for each of the remaining behaviors, a corresponding rule conformance score that indicates the respective behavior's conformity; and

respective values of the rule conformance scores are affected by the order of the evaluation of the conformities;

(vi) based on the rule conformance scores, selecting, by the processor system, one or more of the remaining behaviors, the selection thereby being according to the relative priorities of the prioritization rules to one another without evaluating trajectory-based metrics, wherein the selection of the one or more behaviors is used to generate, by the processor system, a selected motion trajectory for the vehicle; and

(vii) performing, by the processor system, a control of an automated drive of the vehicle according to the generated trajectory.

2 . The method according to claim 1 , wherein, by applying the decision-making process structure to each of the remaining behaviors of the vehicle, a bitset is generated, which encodes, for each respective one of the behaviors, which of the predefined prioritization rules are satisfied, with the remaining behaviors being prioritized based on the generated bitsets.

3 . The method according to claim 1 , wherein, by applying the decision-making process structure for each of the remaining behaviors, a binary decision diagram is generated, with the remaining behaviors being prioritized based on the generated binary decision diagrams.

4 . The method according to claim 1 , wherein, by applying the decision-making process structure for each of the remaining behaviors, a finite decision diagram is generated, with the remaining behaviors being prioritized based on the generated finite decision diagrams.

5 . The method according to claim 1 , wherein the current driving situation is preclassified based on the environmental model by assigning the current driving situation to at least one situation class from a set of specified situation classes, and, when prioritizing the remaining behaviors, the decision-making process structure is limited to those of the prioritization rules that can be applied to situations of the at least one situation class ascertained in this way.

6 . The method according to claim 5 , wherein a context decision tree is used to preclassify the current driving situation.

7 . The method according to claim 1 , wherein the predefined set of prioritization rules includes safety requirements and/or traffic rules and/or comfort requirements and/or technical vehicle boundary conditions.

8 . The method according to claim 1 , wherein at least one of the possible behaviors of the vehicle in the given situation is defined by:

at least a portion of the boundary conditions determined based on the environmental model, or

a set of behavior instructions that implement at least a portion of the boundary conditions determined based on the environmental model.

9 . The method according to claim 1 , further comprising:

subsequent to the selected motion trajectory being generated by the selection of the one or more of the remaining behaviors, optimizing at least a part of the selected motion trajectory with respect to a specified quality function.

10 . A computer-implemented system for controlling a vehicle, the system comprising:

one or more sensors; and

a processor system that includes at least one processor, the processor system being configured to:

(i) obtain a plurality of real-time sensor signals from one or more sensors of the vehicle sensing surroundings of the vehicle, thereby generating aggregated situation-specific information;

(ii) generate an environmental model of a current driving situation of the vehicle based on the aggregated situation-specific information, wherein the environmental model comprises a representation of physical surroundings relevant to the vehicle's operational context;

(iii) determine, based on the environmental model and a predefined set of one or more mandatory compliance rules, one or more boundary conditions defining mandatory constraints on permissible behaviors of the vehicle in the current driving situation of the vehicle;

(iv) based on the determined boundary conditions, eliminate one or more of a plurality of possible behaviors from consideration in a subsequent behavior selection process;

(v) subsequent to the elimination, perform the behavior selection process by applying a decision-making process structure comprising a predefined set of prioritization rules to remaining behaviors of the plurality of possible behaviors which have not been eliminated by the elimination step, wherein:

the decision-making process structure comprises a hierarchical sequence of rule evaluations characterized by that the application of the decision-making process structure evaluates a conformity of the remaining behaviors to the predefined set of prioritization rules in an order that is defined by the hierarchical sequence according to relative priorities of the prioritization rules to one another, resulting in, for each of the remaining behaviors, a corresponding rule conformance score that indicates the respective behavior's conformity; and

respective values of the rule conformance scores are affected by the order of the evaluation of the conformities;

(vi) based on the rule conformance scores, select one or more of the remaining behaviors, the selection thereby being according to the relative priorities of the prioritization rules to one another without evaluating trajectory-based metrics, wherein the selection of the one or more behaviors is used to generate, by the processor system, a selected motion trajectory for the vehicle; and

(vii) perform a control of an automated drive of the vehicle according to the generated trajectory.

11 . The system according to claim 10 , wherein the processor system is configured to perform a preclassification of the current driving situation based on the environmental model and by taking a set of specified situation classes as a basis.

12 . The system according to claim 11 , wherein, as a function of the preclassification of the given situation, different parts of the predefined set of prioritization rules can be activated and deactivated.

13 . The method according to claim 1 , wherein the evaluation of the conformity of the remaining behaviors to the predefined set of prioritization rules is performed without the generation of the motion trajectory that will be using the decision-making process structure.