Obstacle detection systems and methods

• US 10,195,992 B2
• Filed: 04/03/2017
• Issued: 02/05/2019
• Est. Priority Date: 04/03/2017
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Gerardo Castro, Marcos Paul, Liu, Dongran, Kadetotad, Sneha, Jain, Jinesh J
• Primary Examiner(s): Elchanti, Hussein
• Application Number:
  US15/478,118
• Publication Number:
  US 20180281680A1
• Time in Patent Office: 673 Days
• CPC Class Codes: B60R 2300/802 for monitoring and displaying vehicle exterior blind spot views
• CPC Class Codes: B60R 2300/8093 for obstacle warning
• CPC Class Codes: B60W 50/14 Means for informing the driver, warning the driver or prompting a driver intervention
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: G01S 2013/9317 Driving backwards
• CPC Class Codes: G01S 7/4808 Evaluating distance, position or velocity data
• CPC Class Codes: G06F 17/11 for solving equations , e.g. nonlinear equations, general mathematical optimization problems optimization specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Example obstacle detection systems and methods are described. In one implementation, a method receives data from at least one sensor mounted to a vehicle and creates a probabilistic grid-based map associated with an area near the vehicle. The method also determines a confidence associated with each probability in the grid-based map and determines a likelihood that an obstacle exists in the area near the vehicle based on the probabilistic grid-based map.

Claims

1. A method comprising:

receiving, by a controller of a vehicle, data from at least one sensor mounted to the vehicle;

creating, by an obstacle detection system executed by the controller, a probabilistic grid-based map associated with an area near the vehicle, the grid-based map defined as a two-dimensional array of cells distributed in a horizontal plane adjacent the vehicle;

determining, by the obstacle detection system, a confidence associated with each cell of the array of cells in the probabilistic grid-based map; and

determining, by the obstacle detection system, a likelihood that an obstacle exists in the area near the vehicle based on the probabilistic grid-based map;

performing temporal analysis of the received data using learning and forgetting factors associated with the probabilistic grid-based map;

wherein the learning factor increases a probability value associated with each cell in the array of cells over time; and

wherein the forgetting factor decreases a probability value associated with each cell in the array of cells over time.

2. The method of claim 1, wherein the at least one sensor includes at least one of a LIDAR sensor, a radar sensor, and a camera.

3. The method of claim 1, further comprising, responsive to determining a likelihood that an obstacle exists in the area near the vehicle, generating an alert to the driver of the vehicle.

4. The method of claim 1, further comprising, responsive to determining a likelihood that an obstacle exists in the area near the vehicle, communicating an alert to an automated driving system of the vehicle.

5. The method of claim 1, further comprising calculating probabilities associated with neighboring cells in the probabilistic grid-based map based on the received data.

6. The method of claim 5, wherein the neighboring cells include eight cells immediately surrounding a particular cell in the probabilistic grid-based map.

7. The method of claim 5, further comprising updating probabilistic values of the neighboring cells using a Bayesian method.

8. The method of claim 1, wherein the vehicle is an autonomous vehicle.