DE-102024139852-A1 - DETECTION OF UNCOVERED OBJECTS

Abstract

A system for monitoring a vehicle's environment comprises a perception system that communicates with a vehicle sensor, the perception system being designed to receive perception data from the vehicle sensor. The system also includes a scenario detection module designed to detect a first object based on the perception data and to analyze the behavior of the first object in order to predict whether a detection scenario occurs or will occur, in which case a concealed object is revealed. The system further includes a control module designed to, based on the detection of the concealed object as a revealed object, at least control a vehicle to react to the concealed object and/or present information regarding the concealed object to a user.

Inventors

• Ashraf Abualfellat
• Mohammadali Shahriari
• Amanpal S. Grewal
• Dmitriy Feldman

Assignees

• GM Global Technology Operations LLC

Dates

Publication Date

20260513

Application Date

20241230

Priority Date

20241108

Claims (10)

1. A system for monitoring a vehicle's environment, comprising: a perception system that communicates with a vehicle sensor, wherein the perception system is designed to receive perception data from the vehicle sensor; a scenario detection module designed to detect a first object based on the perception data and to analyze the behavior of the first object in order to predict whether a detection scenario occurs or will occur, thereby revealing a concealed object in the detection scenario; and a control module designed to control a vehicle, based on the detection of the concealed object as a revealed object, in order to react to the concealed object and/or to present information regarding the concealed object to a user.
2. System according Claim 1 , wherein the perception system is designed to operate in a normal detection mode and a modified detection mode, object detection in the normal detection mode is performed on the basis of a normal set of criteria, object detection in the modified detection mode is performed on the basis of a modified set of criteria, and the scenario detection module is designed to cause the perception system to switch from the normal detection mode to the modified detection mode based on the prediction that the detection scenario occurs or will occur.
3. System according Claim 2 , wherein the normal detection mode prescribes that the perception system performs object detection based on a first number of detection modalities, and the modified detection mode prescribes that the perception system performs object detection based on a reduced number of detection modalities, where the reduced number is smaller than the first number.
4. System according Claim 2 , wherein a first delay is assigned to the normal detection mode and a second delay is assigned to the modified detection mode, the second delay being smaller than the first delay.
5. System according Claim 1 , where the scenario detection module is designed to predict, based on a maneuver performed by the first object, that the detection scenario occurs or will occur.
6. System according Claim 5 , where the first object is a vehicle traveling ahead of the vehicle, the concealed object is a third vehicle in front of the vehicle traveling ahead, and the maneuver is an aggressive maneuver performed by the first object.
7. System according Claim 1 , where the scenario detection module is designed to predict, based on classifying behavior according to a machine learning model, that the detection scenario occurs or will occur.
8. System according Claim 7 , where the first object is a preceding vehicle driving in front of the vehicle and the machine learning model is designed to determine whether the behavior is classified as an aggressive maneuver.
9. System according Claim 8 , wherein the perception system is designed to perform object detection according to a normal detection mode and a modified detection mode, wherein the normal detection mode is associated with a first delay, the modified detection mode is associated with a second delay that is shorter than the first delay, and the scenario detection module is designed to cause the perception system to operate in the modified detection mode based on determining that the behavior is classified as an aggressive maneuver.
10. A method for monitoring a vehicle's environment, comprising: Receiving perception data from a vehicle sensor of a vehicle's perception system; Detecting a first object in an environment around the vehicle based on the perception data; Analyzing the behavior of the first object to predict whether a detection scenario occurs or will occur, thereby revealing a concealed object in the detection scenario; and based on the prediction that the detection scenario occurs or will occur, and based on the detection of the concealed object as the revealed object, performing a control of the vehicle to respond to the concealed object and/or presenting information regarding the concealed object to a user.

Description

INTRODUCTION This disclosure relates to the field of vehicle perception. In particular, this disclosure relates to systems and methods for monitoring vehicle environments and for controlling a vehicle perception system based on the detection of detected objects. Vehicles are increasingly equipped with sensors and perception devices that enhance the situational awareness of vehicle control systems and drivers, thereby supporting autonomous driving and/or driver assistance. For example, vehicles can feature autonomous and/or semi-autonomous driving modes, such as fully autonomous driving and automated control of certain functions (e.g., parking assist, automated steering during highway driving, brake assist, etc.). Perception systems should be capable of detecting a wide variety of dynamic situations and objects. Improving aspects of object detection and response to dynamic events is desirable. SUMMARY According to an exemplary embodiment, a system for monitoring a vehicle's environment comprises a perception system that communicates with a vehicle sensor, wherein the perception system is configured to receive perception data from the vehicle sensor. The system also comprises a scenario detection module configured to detect a first object based on the perception data and to analyze the behavior of the first object in order to predict whether a detection scenario occurs or will occur, in which a concealed object is revealed. The system further comprises a control module configured to perform at least one control operation of a vehicle to react to the concealed object and/or to present information regarding the concealed object to a user, based on the detection of the concealed object as the revealed object. In addition to one or more of the features described herein, the perception system is designed to operate in a normal detection mode and a modified detection mode, with object detection in the normal detection mode being performed on the basis of a normal set of criteria and object detection in the modified detection mode being performed on the basis of a modified set of criteria, and the scenario detection module being designed to cause the perception system to switch from the normal detection mode to the modified detection mode based on the prediction that the detection scenario occurs or will occur. In addition to one or more of the features described here, the normal detection mode specifies that the perception system performs object detection based on an initial number of detection modalities, and the modified detection mode specifies that the perception system performs object detection based on a reduced number of detection modalities, the reduced number being smaller than the initial number. In addition to one or more of the features described here, a first delay is assigned to the normal detection mode and a second delay is assigned to the modified detection mode, the second delay being shorter than the first delay. In addition to one or more of the features described here, the scenario detection module is designed to predict, based on a maneuver performed by the first object, that the detection scenario occurs or will occur. In addition to one or more of the features described here, the first object is a vehicle traveling ahead of the vehicle, the concealed object is a third vehicle in front of the vehicle traveling ahead, and the maneuver is an aggressive maneuver performed by the first object. In addition to one or more of the features described here, the scenario detection module is designed to predict, based on a classification of behavior according to a machine learning model, that the detection scenario occurs or will occur. In addition to one or more of the features described here, the first object is a vehicle driving ahead of the vehicle, and the machine learning model is designed to determine whether the behavior is classified as an aggressive maneuver. In addition to one or more of the features described herein, the perception system is designed to perform object detection according to a normal detection mode and a modified detection mode, wherein the normal detection mode is associated with a first delay, the modified detection mode is associated with a second delay that is shorter than the first delay, and the scenario detection module is designed to cause the perception system to operate in the modified detection mode based on determining that the behavior is classified as an aggressive maneuver. According to a further embodiment, a method for monitoring a vehicle's environment comprises: receiving perception data from a vehicle sensor of a vehicle's perception system, detecting a first object in the environment around the vehicle based on the perception data, and analyzing the behavior of the first object to predict whether a detection scenario occurs or will occur, wherein a concealed object is revealed in the detection scenario. The method further comprises, based on the prediction