Method and system for estimating an accident risk of an autonomous vehicle
OwnerWalter Steven Rosenbauminventor
Life of this patent
Estimating the term…
Abstract
A method for estimating an accident risk of an autonomous driving unit produces helpful results with fewer autonomous driving cycles. From driving values, which have been determined from monitoring at least one driving parameter of the driving unit during autonomous driving, an autonomous-driving quantity is determined quantifying an autonomous-driving quality of the driving of the driving unit. The autonomous-driving quantity is associated with a plurality of manual-driving quantities, which have been determined from the same driving parameter during manual driving periods of different driving units. An autonomous-driving accident rate value is determined from accident rate values associated to those manual-driving quantities.
Claims21
A computer-implemented method for controlling an autonomous vehicle, the method comprising:
- monitoring with a sensor in the autonomous vehicle, at least one driving parameter of the autonomous vehicle during autonomous driving and determining driving values;
- by a server, determining an autonomous-driving quality metric which quantifies an autonomous-driving quality of the driving of the autonomous vehicle based on the determined driving values;
- accessing a manual-driving quality metrics database including a plurality of manual-driving quality metrics, which have been determined from the at least one driving parameter during manual driving periods of different autonomous vehicles;
- by the server, pre-selecting, according to one or more restriction parameters, a group of manual-driving quality metrics from the plurality of manual-driving quality metrics stored in the database;
- by the server, selecting metrics from the pre-selected group of manual-driving quality metrics that show a predetermined degree of statistical similarity with the determined autonomous-driving quality metric, wherein the predetermined degree of statistical similarity is tested by Null-Hypothesis;
- by the server, determining an autonomous-driving accident rate value from accident rate values associated with the selected manual-driving quality metrics;
- and
- updating autonomous-driving control software of the autonomous vehicle based on the determined autonomous-driving accident rate value, and controlling the autonomous vehicle by the autonomous-driving control software that was updated.
A system for controlling an autonomous vehicle, the system comprising:
- a sensor system of the autonomous vehicle configured for monitoring a driving parameter during a driving period in which the autonomous vehicle is autonomously driven;
- a server configured to receive, from said sensor system, signals related to the driving parameter;
- and
- autonomous-driving control software of the autonomous vehicle;
- said server configured for determining, from the driving parameter, an autonomous driving quality metric that quantifies an autonomous-driving quality of the driving of the autonomous vehicle based on driving values;
- said server configured for accessing a manual-driving quality metrics database and said server configured for pre-selecting, according to one or more restriction parameters, a group of manual-driving quality metrics from a plurality of manual-driving quality metrics stored in the manual-driving quality metrics database, wherein the plurality of manual-driving quality metrics have been determined from the driving parameter during manual driving periods of different autonomous vehicles;
- said server configured for selecting metrics from the pre-selected group of manual-driving quality metrics that show a predetermined degree of statistical similarity with the determined autonomous-driving quality metric, wherein the predetermined degree of statistical similarity is tested by Null-Hypothesis;
- said server configured for determining an autonomous-driving accident rate value from accident rate values associated with the selected manual-driving quality metrics;
- and
- said autonomous-driving control software of the autonomous vehicle configured for being updated based on the determined autonomous-driving accident rate value and for the controlling the autonomous vehicle after being updated based on the determined autonomous-driving accident rate value.
Dependencies are inferred from the claim text.