Predictive Vehicle Operation Using Worst-Case Route Cost Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive driving strategies for motor vehicles, which rely on future route information, face challenges in ensuring compliance with emission limits and energy efficiency, as they are not guaranteed to work in all situations, leading to increased fuel consumption and emissions, and are not permissible under legislation due to uncertainty in route choice and potential deviations.
Innovation Solution
A method for predictive vehicle operation that uses a digital map and cost function to determine a worst-case scenario within a prediction horizon, allowing for iterative calculation of a predictive driving function that reduces emissions and fuel consumption, even without exact route knowledge, thereby providing a turnkey solution compliant with emissions guidelines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If predictive driving strategies are used based on future route information, then fuel consumption and emissions are reduced, but reliability deteriorates because they are not guaranteed to work in all situations
Solution Approach 1:
The patent applies preliminary action by calculating and storing worst-case cost functions in advance for all possible routes within a prediction horizon. These pre-calculated cost functions represent the most unfavorable scenarios for fuel consumption and emissions. When making driving decisions, the system uses these pre-prepared worst-case scenarios to ensure compliance even under the most adverse conditions, rather than relying on optimistic predictions that may not materialize.
Solution Approach 2:
The patent implements beforehand cushioning by designing the predictive driving strategy to accommodate worst-case scenarios. By planning for the most unfavorable conditions in advance (such as maximum gradient, longest route, highest traffic density), the system creates a safety margin that ensures emission compliance even when actual conditions deviate from predictions. This cushioning approach allows the system to meet emission limits under all possible situations, not just typical conditions.
2Reliability
If conventional strategies are used that work in all situations, then reliability is maintained, but fuel consumption increases significantly
Solution Approach 1:
The patent applies partial or excessive action by using only the necessary level of conservatism required for compliance. Instead of always planning for the absolute worst-case scenario across all possible routes, the system calculates cost functions for multiple possible routes and selects the appropriate level of conservatism based on the specific situation. This allows the system to achieve compliance with emission limits while avoiding the excessive fuel consumption that would result from always assuming the worst possible conditions.
3Use of energy by moving object
If predictive strategies are used with exact route knowledge, then fuel consumption is minimized, but adaptability deteriorates when route deviations occur
Solution Approach 1:
The patent applies dynamics by making the predictive driving strategy adaptable to changing conditions. The system continuously monitors actual route deviations and dynamically adjusts the predictive calculations based on the current situation. When the vehicle deviates from the predicted route, the system recalculates cost functions for the actual route taken, ensuring that fuel optimization continues to be effective even though the original predictions were based on different route information. This dynamic adaptation maintains both fuel efficiency and route flexibility.
Data Source
Figure 1
AI summary
The invention relates to a method for the predictive operation of a motor vehicle based on a predictive driving function, a cost function characteristic of the driving function, and a digital map containing map attributes. The method comprises steps for extracting (S1) map attributes within a prediction horizon around an instantaneous position of the motor vehicle; determining (S2) the cost function characteristic of the predictive driving function; determining (S3) possible routes within the prediction horizon using a navigation algorithm; determining (S4) a maximum or a minimum of the cost function from the possible routes to obtain a worst-case cost function; and calculating (S5) the predictive driving function based on the worst-case cost function or based on the map attributes of the worst-case cost function.