Safety procedure analysis for obstacle avoidance in autonomous vehicles

Patent metadata

Field	Value
Publication number	US-11079764-B2
Application number	US-201916265780-A
Country	US
Kind code	B2
Filing date	Feb 1, 2019
Priority date	Feb 2, 2018
Publication date	Aug 3, 2021
Grant date	Aug 3, 2021

Abstract

In various examples, a current claimed set of points representative of a volume in an environment occupied by a vehicle at a time may be determined. A vehicle-occupied trajectory and at least one object-occupied trajectory may be generated at the time. An intersection between the vehicle-occupied trajectory and an object-occupied trajectory may be determined based at least in part on comparing the vehicle-occupied trajectory to the object-occupied trajectory. Based on the intersection, the vehicle may then execute the first safety procedure or an alternative procedure that, when implemented by the vehicle when the object implements the second safety procedure, is determined to have a lesser likelihood of incurring a collision between the vehicle and the object than the first safety procedure.

First claim

What is claimed is:

1. A processor comprising processing circuitry to:
determine a current claimed set of points representative of a volume in an environment occupied by a vehicle at a time based at least in part on a size, location, and orientation of the vehicle;
based at least in part on one or more velocities of the vehicle at the time and a first safety procedure for the vehicle, generate one or more vehicle-occupied trajectories representative of first future claimed sets of points of the vehicle in the environment if the vehicle were to implement the first safety procedure at the time;
detect an intersection between the one or more vehicle-occupied trajectories and one or more object-occupied trajectories based at least in part on determining whether one or more first vertices of any of the one or more vehicle-occupied trajectories intersects with one or more first edges of any of the one or more object-occupied trajectories or determining whether one or more second vertices of any of the one or more object-occupied trajectories intersects with one or more second edges of any of the one or more vehicle-occupied trajectories, the one or more object-occupied trajectories representative of second future claimed sets of an object in the environment if the object were to implement a second safety procedure associated with the object at the time;
calculate a first safety potential associated with the vehicle implementing the first safety procedure and the object implementing the second safety procedure, the first safety potential representative of the intersection; and
based at least in part on the intersection, execute one of the first safety procedure or an alternative procedure that, when implemented by the vehicle when the object implements the second safety procedure, is determined to have a second safety potential computed to have a lesser likelihood of incurring a collision between the vehicle and the object than the first safety potential.

2. The processor of claim 1 , wherein the processing circuitry is further to determine the one or more velocities, the orientation, and the location of the vehicle based at least in part on sensor data received from one or more sensors of the vehicle, the one or more sensors including one or more of:
a global navigation satellite system (GNSS) sensor;
a camera;
an accelerometer;
an inertial sensor;
a gyrosensor;
a compass;
a tire vibration sensor;
a microphone;
a steering sensor; or
a speed sensor.

3. The processor of claim 1 , wherein the processing circuitry is further to determine a set of functions corresponding to a control policy representative of the first safety procedure for the vehicle by:
determining control parameters for steering, braking, and accelerating the vehicle;
identifying a vehicle state objective of the first safety procedure;
analyzing sensor data received from one or more sensors of the vehicle to determine at least one of locations, orientations, and velocities of one or more objects in the environment; and
based at least in part on the control parameters and at least one of the locations, the orientations, and the velocities of the one or more objects, determining the set of functions to guide the vehicle to the vehicle state objective.

4. The processor of claim 1 , wherein the processing circuitry is further to determine the size of the vehicle by:
identifying an estimated size of the vehicle; and
fitting a predefined shape around the vehicle based at least in part on at least one of the estimated size or a vehicle type of the vehicle,
wherein the size of the vehicle corresponds to a size of the predefined shape.

5. The processor of claim 4 , wherein the determining the size of the vehicle further comprises identifying a safety margin, and wherein the predefined shape includes the safety margin.

6. The method processor of claim 1 , wherein the processing circuitry is further to determine the one or more object-occupied trajectories for the object by determining at least one of an object velocity, an object orientation, and at least one of an object location, an object size, or an object shape of the object in the environment based at least in part on sensor data received from one or more sensors of the vehicle, the one or more sensors including one or more of:
a global navigation satellite system (GNSS) sensor;
a long-range camera;
a stereo camera;
an infrared camera;
a surround camera;
a wide view camera;
a mid-range camera;
a LIDAR sensor;
an ultrasonic sensor;
an infrared sensor;
a radar sensor; or
a wireless antenna.

7. The processor of claim 1 , wherein each first future claimed set of the one or more first future claimed sets is representative of a respective volume in the environment that would be occupied by the vehicle at a respective future time.

8. A method comprising:
determining a current claimed set of points representative of an area in an environment occupied by a vehicle at a time based at least in part on a state of the vehicle, the state including at least a location and a speed of the vehicle;
based at least in part on the state and a safety procedure for the vehicle, generating a vehicle-occupied trajectory representative of future claimed sets of the vehicle in the environment if the vehicle were to implement the safety procedure at the time;
comparing the vehicle-occupied trajectory to one or more object-occupied trajectories associated with one or more objects in the environment to detect whether any virtual intersections between the vehicle-occupied trajectory and at least one of the one or more object-occupied trajectories occur, the comparing including at least one of determining whether one or more first vertices of the vehicle-occupied trajectory intersects with one or more first edges of any of the one or more object-occupied trajectories or determining whether one or more second vertices of any of the one or more object-occupied trajectories intersects with one or more second edges of the vehicle-occupied trajectory; and
upon detecting a virtual intersection, implementing a first set of controls defined by the safety procedure or a second set of controls defined by an alternative procedure determined to have a lesser likelihood, compared to the safety procedure, of incurring a real-world intersection between the vehicle and at least one of the one or more objects associated with the virtual intersection.

9. The method of claim 8 , wherein the implementing the first set of controls or the second set of controls comprises:
calculating a first safety potential representative of a first likelihood of the real-world intersection based at least in part on the first set of controls representative of the safety procedure;
calculating a second safety potential representative of a second likelihood of the real-world intersection based at least in part on the second set of controls representative of the alternative procedure;
comparing the first safety potential to the second safety potential; and
based at least in part on the comparing:
upon determining that the second likelihood is greater than the first likelihood, implementing the first set of controls; or
upon determining that the second likelihood is equal to or less than the first likelihood, implementing the first set of controls or the second set of controls.

10. The method of claim 8 , wherein the time is a current time plus an estimated latency, the estimated latency being predefined or determined based at least in part on historical data associated with the vehicle.

11. The method of claim 8 , further comprising generating the one or more object-occupied trajectories by:
based at l

Assignees

• Nvidia Corp

Inventors

• Nister David
• Lee Hon-Leung
• Ng Julia
• Wang Yizhou

Classifications

• B60W2554/80

  Spatial relation or speed relative to objects · CPC title

• B60W60/0015

  specially adapted for safety · CPC title

• B60W2554/4026

  Cycles · CPC title

• B60W2554/4029

  Pedestrians · CPC title

• G05D1/247

  using signals provided by artificial sources external to the vehicle, e.g. navigation beacons · CPC title