Potential collision warning system based on road user intent prediction

Patent metadata

Field	Value
Publication number	US-11345342-B2
Application number	US-201916586665-A
Country	US
Kind code	B2
Filing date	Sep 27, 2019
Priority date	Sep 27, 2019
Publication date	May 31, 2022
Grant date	May 31, 2022

Abstract

An apparatus comprising a memory to store an observed trajectory of a pedestrian, the observed trajectory comprising a plurality of observed locations of the pedestrian over a first plurality of timesteps; and a processor to generate a predicted trajectory of the pedestrian, the predicted trajectory comprising a plurality of predicted locations of the pedestrian over the first plurality of timesteps and over a second plurality of timesteps occurring after the first plurality of timesteps; determine a likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the first plurality of timesteps and the plurality of observed locations of the pedestrian over the first plurality of timesteps; and responsive to the determined likelihood of the predicted trajectory, provide information associated with the predicted trajectory to a vehicle to warn the vehicle of a potential collision with the pedestrian.

First claim

What is claimed is:

1. An apparatus comprising:
a memory to store a first observed trajectory and a second observed trajectory of a pedestrian, the first observed trajectory comprising a plurality of observed locations of the pedestrian over a first plurality of timesteps, the second observed trajectory comprising a plurality of observed locations of the pedestrian over a second plurality of timesteps, wherein the second plurality of timesteps includes the first plurality of timesteps and at least one additional subsequent timestep; and
a processor coupled to the memory, the processor comprising circuitry, the processor to:
generate a predicted trajectory of the pedestrian, the predicted trajectory comprising a plurality of predicted locations of the pedestrian over the second plurality of timesteps and over a third plurality of timesteps occurring after the second plurality of timesteps;
determine a first likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the first plurality of timesteps and the plurality of observed locations of the pedestrian over the first plurality of timesteps;
determine a second likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the second plurality of timesteps and the plurality of observed locations of the pedestrian over the second plurality of timesteps; and
responsive to the first and/or second likelihood of the predicted trajectory, provide information associated with the predicted trajectory to a vehicle to warn the vehicle of a potential collision with the pedestrian.

2. The apparatus of claim 1 , wherein the vehicle does not have a line of sight with the pedestrian when the information associated with the predicted trajectory is provided by the processor.

3. The apparatus of claim 1 , wherein the information associated with the predicted trajectory comprises the predicted trajectory.

4. The apparatus of claim 1 , wherein the processor is further to select the predicted trajectory from among a plurality of predicted trajectories for the pedestrian based on the determined likelihood of the predicted trajectory.

5. The apparatus of claim 1 , wherein the comparison of the plurality of predicted locations of the pedestrian over the first plurality of timesteps and the plurality of observed locations of the pedestrian over the first plurality of timesteps comprises an evaluation of values of a plurality of gaussian distributions at the plurality of predicted locations of the pedestrian over the first plurality of timesteps, wherein the plurality of gaussian distributions have respective means at the plurality of observed locations of the pedestrian over the first plurality of timesteps.

6. The apparatus of claim 1 , wherein the processor is further to generate a path from a start position to an inferred goal position of the pedestrian using random sampling and generate the predicted trajectory of the pedestrian based on the generated path.

7. The apparatus of claim 6 , wherein the random sampling is generated utilizing a Rapidly-exploring Random Tree Star.

8. The apparatus of claim 1 , wherein the processor is further to:
determine locations of a plurality of obstacles of an intersection, an obstacle representing an area of the intersection that is not traversable by the pedestrian; and
generate the predicted trajectory of the pedestrian based on the plurality of obstacles of the intersection.

9. The apparatus of claim 1 , wherein the determination of the second likelihood of the predicted trajectory is further based on historical information indicating how often pedestrians travel to a goal position of the predicted trajectory relative to other goal positions.

10. The apparatus of claim 1 , wherein the predicted trajectory of the pedestrian is based at least in part on an observed motion profile of the pedestrian.

11. The apparatus of claim 1 , wherein the predicted trajectory of the pedestrian is based at least in part on a state of a traffic light of an intersection.

12. The apparatus of claim 1 , further comprising one or more of a battery communicatively coupled to the processor, a display communicatively coupled to the processor, or a network interface communicatively coupled to the processor.

13. The apparatus of claim 1 , wherein the processor is to generate a plurality of predicted trajectories of the pedestrian from a start position to an inferred goal position based on different assumed velocities or accelerations of the pedestrian, wherein the predicted trajectory is one of the plurality of predicted trajectories.

14. A method comprising:
storing first observed trajectory and a second observed trajectory of a pedestrian, the first observed trajectory comprising a plurality of observed locations of the pedestrian over a first plurality of timesteps, the second observed trajectory comprising a plurality of observed locations of the pedestrian over a second plurality of timesteps, wherein the second plurality of timesteps includes the first plurality of timesteps and at least one additional subsequent timestep;
generating a predicted trajectory of the pedestrian, the predicted trajectory comprising a plurality of predicted locations of the pedestrian over the second plurality of timesteps and over a third plurality of timesteps occurring after the second plurality of timesteps;
determining a first likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the first plurality of timesteps and the plurality of observed locations of the pedestrian over the first plurality of timesteps;
determining a second likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the second plurality of timesteps and the plurality of observed locations of the pedestrian over the second plurality of timesteps; and
responsive to the first and/or second likelihood of the predicted trajectory, providing information associated with the predicted trajectory to a vehicle to warn the vehicle of a potential collision with the pedestrian.

15. The method of claim 14 , wherein the vehicle does not have a line of sight with the pedestrian when the information associated with the predicted trajectory is provided.

16. The method of claim 14 , further comprising generating a path from a start position to an inferred goal position of the pedestrian using random sampling and generating the predicted trajectory of the pedestrian based on the generated path.

17. The method of claim 14 , further comprising:
determining locations of a plurality of obstacles of an intersection, an obstacle representing an area of the intersection that is not traversable by the pedestrian; and
generating the predicted trajectory of the pedestrian based on the plurality of obstacles of the intersection.

18. At least one non-transitory machine readable storage medium having instructions stored thereon, the instructions when executed by a machine to cause the machine to:
store a first observed trajectory and a second observed trajectory of a pedestrian, the first observed trajectory comprising a plurality of observed locations of the pedestrian over a first plurality of timesteps, the second observed trajectory comprising a plurality of observed locations of the pedestrian over a second plurality of timesteps, wherein the second plurality of timesteps includes the first plurality of timesteps and at least o

Assignees

• Intel Corp

Inventors

• Gutierrez David Gomez
• Felip Leon Javier
• Doshi Kshitij A
• Campos Macias Leobardo E
• Ahuja Nilesh Amar
• Tickoo Omesh

Classifications

• B60W2554/4026

  Cycles · CPC title

• B60W2554/4029

  Pedestrians · CPC title

• G06V20/584

  of vehicle lights or traffic lights · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V10/62

  relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking · CPC title