Autonomous vehicle computing system with processing assurance

Patent metadata

Field	Value
Publication number	US-11458997-B2
Application number	US-202016893617-A
Country	US
Kind code	B2
Filing date	Jun 5, 2020
Priority date	Mar 31, 2020
Publication date	Oct 4, 2022
Grant date	Oct 4, 2022

Abstract

Systems and methods are directed to a method for assured autonomous vehicle compute processing. The method can include providing sensor data to first and second functional circuitry of an autonomy computing system. The first and second functional circuitry can be configured to generate first and second outputs associated with a first autonomous compute function. The method can include generating, by the first and second functional circuitry in response to the sensor data, first and second output data associated with the first autonomous compute function. The method can include generating, by monitoring circuitry of the autonomy computing system, comparative data associated with differences between the first output data and the second output data. The method can include generating one or more vehicle control signals for the autonomous vehicle based at least in part on the comparative data.

First claim

What is claimed is:

1. A computer-implemented method for assured autonomous vehicle compute processing, comprising:
providing data associated with a sensor system of an autonomous vehicle to first functional circuitry and second functional circuitry of an autonomy computing system of a vehicle computing system, the first functional circuitry configured to generate one or more first outputs associated with a first autonomous compute function of the autonomy computing system and the second functional circuitry configured to generate one or more second outputs associated with the first autonomous compute function of the autonomy computing system;
generating, by the first functional circuitry in response to the data associated with the sensor system, first output data associated with the first autonomous compute function of the autonomy computing system;
generating, by the second functional circuitry in response to the data associated with the sensor system, second output data associated with the first autonomous compute function of the autonomy computing system;
generating, by monitoring circuitry of the autonomy computing system, comparative data associated with one or more differences between the first output data associated with the first autonomous compute function of the autonomy computing system and the second output data associated with the first autonomous compute function of the autonomy computing system; and
generating, by a vehicle computing system, one or more vehicle control signals for the autonomous vehicle based at least in part on the comparative data associated with the one or more differences between the first output data and the second output data.

2. The computer-implemented method of claim 1 , wherein:
the first functional circuitry is configured to generate the first output data associated with the first autonomous compute function of the autonomy computing system at a first frequency; and
the second functional circuitry is configured to generate the second output data associated with the first autonomous compute function of the autonomy computing system at a second frequency that is less than the first frequency.

3. The computer-implemented method of claim 1 , wherein:
the first functional circuitry includes one or more non-assured hardware processing circuits and the second functional circuitry includes one or more non-assured processing circuits; and
the monitoring circuitry includes one or more assured processing circuits.

4. The computer-implemented method of claim 1 , wherein:
the first functional circuitry includes one or more assured hardware processing circuits and the second functional circuitry includes one or more assured processing circuits; and
the method further comprises monitoring, by the monitoring circuitry, performance of one or more operations associated with at least one of the first functional circuitry or the second functional circuitry.

5. The computer-implemented method of claim 4 , wherein
the one or more first outputs of the first functional circuitry is one or more first non-assured outputs;
the one or more second outputs of the second functional circuitry is one or more second non-assured outputs; and
the method further comprises providing, by the monitoring circuitry, one or more assured outputs associated with the one or more first non-assured outputs and the one or more second non-assured outputs.

6. The computer-implemented method of claim 1 , wherein each of the first functional circuitry and the second functional circuitry respectively comprise:
a first processor and a second processor;
a first core of a processor and a second core of the processor; or
a first computing device and a second computing device.

7. The computer-implemented method of claim 1 , wherein the monitoring circuitry comprises virtualized processing circuitry.

8. The computer-implemented method of claim 1 , wherein generating, by the monitoring circuitry of the autonomy computing system, the comparative data associated with the one or more differences between the first output data associated with the first autonomous compute function of the autonomy computing system and the second output data associated with the first autonomous compute function of the autonomy computing system comprises:
using, by the autonomy computing system, the first functional circuitry to generate a second output validation for the one or more second outputs;
using, by the autonomy computing system, the second functional circuitry to generate a first output validation for the one or more first outputs; and
generating, by the monitoring circuitry of the autonomy computing system, the comparative data based at least in part on the first output validation and the second output validation.

9. The computer-implemented method of claim 8 , wherein:
Using, by the autonomy computing system, the first functional circuitry to generate the second output validation for the one or more second outputs comprises inputting the one or more second outputs to one or more first neural networks associated with the first functional circuitry to generate a second output validation; and
using, by the autonomy computing system, the second functional circuitry to generate the first output validation for the one or more first outputs comprises inputting the one or more first outputs to one or more second neural networks associated with the second functional circuitry to generate a first output validation.

10. The computer-implemented method of claim 8 , wherein:
the comparative data indicates a fault associated with the one or more first outputs; and
generating the one or more vehicle control signals comprises generating the one or more vehicle control signals based at least in part on the second output data.

11. The computer-implemented method of claim 9 , further comprising:
controlling, by the vehicle computing system, the autonomous vehicle based at least in part on the one or more vehicle control signals.

12. The computer-implemented method of claim 1 , wherein:
generating, by the first functional circuitry in response to the data associated with the sensor system, the first output data comprises using, by the first functional circuitry, one or more first neural networks to generate the first output data; and
generating, by the second functional circuitry in response to the data associated with the sensor system, the second output data comprises using, by the second functional circuitry, one or more second neural networks to generate the second output data.

13. The computer-implemented method of claim 1 , further comprising:
evaluating at least one output of a perception system of the autonomy computing system using a perception checking system;
evaluating at least one output of a prediction system of the autonomy computing system using a prediction checking system; and
evaluating at least one output of a motion planning system of the autonomy computing system using a motion planning checking system.

14. The computer-implemented method of claim 1 , wherein:
the first output data comprises perception data associated with a first object; and
the second output data comprises perception data associated with the first object.

15. The computer-implemented method of claim 1 , wherein:
the first output data includes motion planning data comprising a trajectory for the autonomous vehicle; and
the second output data includes motion planning data comprising the trajectory for the autonomous vehicle.

16. An autonomy computing system for an autonomous vehicle, comprising:
first func

Assignees

• Uatc Llc

Inventors

• Hyde Sean
• Molinari Jose Francisco
• Thomas Stephen Luke

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• B60W60/0027

  using trajectory prediction for other traffic participants · CPC title

• G06N3/08

  Learning methods · CPC title