Autonomous vehicle routing based upon risk of autonomous vehicle takeover

Patent metadata

Field	Value
Publication number	US-11112794-B2
Application number	US-201916280599-A
Country	US
Kind code	B2
Filing date	Feb 20, 2019
Priority date	Feb 20, 2019
Publication date	Sep 7, 2021
Grant date	Sep 7, 2021

Abstract

Various technologies described herein pertain to routing an autonomous vehicle based upon risk of takeover of the autonomous vehicle by a human operator. A computing system receives an origin location and a destination location of the autonomous vehicle. The computing system identifies a route for the autonomous vehicle to follow from the origin location to the destination location based upon output of a computer-implemented model. The computer-implemented model is generated based upon labeled data indicative of instances in which autonomous vehicles are observed to transition from operating autonomously to operating based upon conduction by human operators while the autonomous vehicles are executing predefined maneuvers. The computer-implemented model takes, as input, an indication of a maneuver in the predefined maneuvers that is performed by the autonomous vehicle when the autonomous vehicle follows a candidate route. The autonomous vehicle then follows the route from the origin location to the destination location.

First claim

What is claimed is:

1. A computing system comprising:
a processor; and
non-transitory memory that stores computer-readable instructions that, when executed by the processor, cause the processor to perform acts comprising:
receiving an origin location of an autonomous vehicle and a destination location of the autonomous vehicle; and
identifying a route for the autonomous vehicle to follow from the origin location to the destination location based upon output of a computer-implemented model, wherein the computing system identifies the route from amongst a plurality of candidate routes, wherein the computer-implemented model is generated based upon labeled data that is indicative of instances in which autonomous vehicles are observed to transition from operating autonomously to operating based upon conduction by human operators while the autonomous vehicles are executing predefined maneuvers, wherein the computer-implemented model takes, as input, an indication of a maneuver in the predefined maneuvers that is performed by the autonomous vehicle when the autonomous vehicle follows a candidate route in the candidate routes, wherein the output of the computer-implemented model is a score that is indicative of a likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by a human operator due to the autonomous vehicle executing the maneuver along the candidate route, wherein the score is used in part to identify the route from amongst the candidate routes, and wherein the autonomous vehicle follows the route from the origin location to the destination location.

2. The computing system of claim 1 , wherein the predefined maneuvers include one or more of:
remaining in a lane on a road;
a left lane change;
a right lane change;
a left turn;
a right turn; or
remaining stationary.

3. The computing system of claim 1 , wherein the computer-implemented model comprises one of:
a mixed model;
a Bayesian hierarchical model;
a random forest model; or
a neural network.

4. The computing system of claim 1 , wherein the computing system is comprised by the autonomous vehicle, wherein the autonomous vehicle further comprises:
a vehicle propulsion system;
a braking system; and
a steering system,
wherein the autonomous vehicle controls at least one of the vehicle propulsion system, the braking system, or the steering system in order to follow the route from the origin location to the destination location.

5. The computing system of claim 1 , the acts further comprising:
prior to identifying the route for the autonomous vehicle to follow from the origin location to the destination location based upon the output of the computer-implemented model, receiving the labeled data; and
generating the computer-implemented model based on the labeled data.

6. The computing system of claim 1 , wherein the candidate route is identified based on travel time from the origin location to the destination location being minimized, and wherein the route is identified based on the likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by the human operator being minimized.

7. The computing system of claim 1 , wherein the computing system generates a weighted directed graph that represents the candidate routes, wherein the weighted directed graph comprises nodes and directed edges coupling at least some of the nodes, wherein the nodes represent intersections in a driving environment of the autonomous vehicle, wherein the driving environment includes the origin location and the destination location, wherein the directed edges represent roads that connect at least some of the intersections, and wherein weights assigned to the directed edges are indicative of costs to the autonomous vehicle for traversing the intersections.

8. The computing system of claim 7 , wherein the computing system updates a weight in the weights in the weighted directed graph based upon the score, and wherein the score is indicative of the likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by the human operator due to the autonomous vehicle executing the maneuver at a particular intersection represented by one of the nodes in the weighted directed graph.

9. The computing system of claim 8 , wherein the computing system identifies the route by applying a shortest path algorithm to the weighted directed graph.

10. The computing system of claim 9 , wherein the shortest path algorithm is one of a Dijkstra's algorithm, a Bellman-Ford algorithm, or a Floyd-Warshall algorithm.

11. The computing system of claim 1 , wherein the computer-implemented model further takes, as input, an indication of a weather condition, and wherein the score that is outputted by the computer-implemented model is indicative of the likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by the human operator due to the autonomous vehicle executing the maneuver along the candidate route in the weather condition.

12. The computing system of claim 1 , wherein the computer-implemented model further takes, as input, an indication of a time of day, and wherein the score that is outputted by the computer-implemented model is indicative of the likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by the human operator due to the autonomous vehicle executing the maneuver along the candidate route at the time of day.

13. A method executed by a processor of a computing system, the method comprising:
receiving an origin location of an autonomous vehicle and a destination location of the autonomous vehicle;
identifying a route for the autonomous vehicle to follow from the origin location to the destination location based upon output of a computer-implemented model, wherein the computing system identifies the route from amongst a plurality of candidate routes, wherein the computer-implemented model is generated based upon labeled data that is indicative of instances in which autonomous vehicles are observed to transition from operating autonomously to operating based upon conduction by human operators while the autonomous vehicles are executing predefined maneuvers, wherein the computer-implemented model takes, as input, an indication of a maneuver in the predefined maneuvers that is performed by the autonomous vehicle when the autonomous vehicle follows a candidate route in the candidate routes, wherein the output of the computer-implemented model is a score that is indicative of a likelihood that the autonomous vehicle will be caused to transition from operating autonomously to operating based upon conduction by a human operator due to the autonomous vehicle executing the maneuver along the candidate route, and wherein the score is used in part to identify the route from amongst the candidate routes; and
transmitting the route to the autonomous vehicle, wherein the autonomous vehicle follows the route from the origin location to the destination location.

14. The method of claim 13 , wherein the predefined maneuvers include one or more of:
remaining in a lane on a road;
a left lane change;
a right lane change;
a left turn at an intersection having a traffic light, a stop sign, or a yield sign;
a right turn at the intersection having the traffic light, the stop sign, or the yield sign; or
remaining stationary.

15. The method of cla

Assignees

• Gm Cruise Holdings Llc

Inventors

• Chi-Johnston Geoffrey Louis
• Vaingankar Vishal Suresh
• Joseph Antony
• Skwerer Sean Gregory
• Rech Lucio Otavio Marchioro
• Passa Nitin Kumar
• Freeman Laura Athena
• Hines George Herbert

Classifications

• B60W60/0021

  specially adapted for travel time · CPC title

• B60W60/0011 Primary

  involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles · CPC title

• G01C21/20

  Instruments for performing navigational calculations (G01C21/24, G01C21/26 take precedence) · CPC title

• G01C21/3453

  Special cost functions, i.e. other than distance or default speed limit of road segments · CPC title

• B60W60/0053

  from vehicle to occupant · CPC title