Method and apparatus for autonomous system performance and grading
The present application generally relates to methods and apparatus for evaluating and assigning a performance metric to a driver response to a driving scenario. More specifically, the application teaches a method and apparatus for breaking a scenario into features, assigning each feature a grade and generating an overall grade in response to a weighted combination of the grades.
1. An apparatus comprising:
a sensor interface for generating sensor data for coupling to a vehicle control system;
a control system interface for receiving a control data from the vehicle control system;
a memory for storing a driving scenario wherein the driving scenario has a first time parsed subtask and a second time parsed subtask wherein the first time parsed subtask has a first complexity and the second time parsed subtask has a second complexity; and
a processor for comparing the control data to the driving scenario, assigning a first grade to the first time parsed subtask and a second grade to the second time parsed subtask, weighting the first grade to generate a first weighted grade and weighting the second grade to generate a second weighted grade, and generating a performance metric based on the first weighted grade, the second weighted grade, the first complexity and the second complexity.
2. The apparatus of claim 1 wherein the performance metric is indicative of a driver performance compared to an optimal performance of the scenario.
3. The apparatus of claim 1 wherein the first time parsed subtask is at least one of stopping velocity, velocity weighted mean distance from an ideal path, post left turn matching of traffic speed, mean distance from other cars, surprise and critical error.
4. The apparatus of claim 1 wherein the control data is generated by an autonomous driving control system.
5. The apparatus of claim 4 wherein the performance metric is transmitted to the autonomous driving control system.
6. The apparatus of claim 1 wherein the performance metric is generated in response to a critical error factor, the first weighted grade and the second weighted grade.
7. A method of driver performance evaluation comprising:
receiving a control data indicative of a driver performance;
comparing the control data to a driving scenario, wherein the driving scenario has a first time parsed subtask and a second time parsed subtask wherein the first time parsed subtask has a first complexity and the second time parsed subtask has a second complexity;
assigning a first grade to the first time parsed subtask and a second grade to the second time parsed subtask;
weighting the first grade to generate a first weighted grade and weighting the second grade to generate a second weighted grade; and
generating a performance metric based on the first weighted grade, the second weighted grade, the first complexity and the second complexity.
8. The method of claim 7 wherein the performance metric is indicative of the driver performance compared to an optimal performance of the scenario.
9. The method of claim 7 wherein the first time parsed subtask is at least one of stopping velocity, velocity weighted mean distance from an ideal path, post left turn matching of traffic speed, mean distance from other cars, surprise and critical error.
10. The method of claim 7 wherein the control data is generated by an autonomous driving control system.
11. The method of claim 10 wherein the performance metric is transmitted to the autonomous driving control system.
12. The method of claim 7 wherein the performance metric is generated in response to a critical error factor, the first weighted grade and the second weighted grade.