Scenario Aware Perception System For An Automated Vehicle

Abstract

A scenario aware perception system suitable for use on an automated vehicle includes a traffic-scenario detector, an object-detection device, and a controller. The traffic-scenario detector is used to detect a present-scenario experienced by a host-vehicle. The object-detection device is used to detect an object proximate to the host-vehicle. The controller is in communication with the traffic-scenario detector and the object-detection device. The controller configured to determine a preferred-algorithm used to identify the object. The preferred-algorithm is determined based on the present-scenario.

Claims

1. A scenario aware perception system suitable for use on an automated vehicle, said system comprising: a traffic-scenario detector used to detect a present-scenario experienced by a host-vehicle; an object-detection device used to detect an object proximate to the host-vehicle; and a controller in communication with the traffic-scenario detector and the object-detection device, said controller configured to determine a preferred-algorithm used to identify the object, wherein the preferred-algorithm is determined based on the present-scenario.

2. The system in accordance with claim 1, wherein the traffic-scenario detector includes a location-indicator, and the present-scenario is determined based on a map-location of the host-vehicle on a digital-map indicated by the location-indicator.

3. The system in accordance with claim 1, wherein the traffic-scenario detector includes one or more of a camera, a lidar-unit, and a radar-sensor, and the present-scenario is determined based on a signal from the traffic-scenario detector.

4. The system in accordance with claim 1, wherein the system includes a memory, and the controller is configured to select the present-scenario from a plurality of possible-scenarios stored in the memory.

5. The system in accordance with claim 1, wherein the object-detection device includes one or more of a camera, a lidar-unit, and a radar-sensor, and the object is detected based on a signal from the object-detection device.

6. The system in accordance with claim 1, wherein the controller is configured to further determine the preferred-algorithm based on an expected-direction of motion of the object relative to the host-vehicle indicated by the present-scenario.

7. The system in accordance with claim 1, wherein the system includes a memory, and the preferred-algorithm is used to process a signal from the object-detection device is selected from a plurality of optimized-algorithms stored in the memory.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0012] The present invention will now be described, by way of example with reference to the accompanying drawings, in which:

[0013] FIG. 1 is a diagram of a scenario aware perception system in accordance with one embodiment;

[0014] FIG. 2 is traffic scenario encountered by the system of FIG. 1 in accordance with one embodiment; and

[0015] FIG. 3 is traffic scenario encountered by the system of FIG. 1 in accordance with one embodiment.

DETAILED DESCRIPTION

[0016] FIG. 1 illustrates a non-limiting example of a scenario aware perception system 10, hereafter referred to as the system 10, which is generally suitable for use on an automated vehicle, hereafter referred to as the host-vehicle 12. The system 10 includes a traffic-scenario detector 14 used to determine or detect a present-scenario 16, i.e. a traffic-situation, presently being experienced by the host-vehicle 12. As used herein, the present-scenario 16 may be characterized by the configuration of a roadway 18 proximate to the host-vehicle 12. By way of example and not limitation, the present-scenario 16 may be characterized as a multilane expressway (FIG. 2), a two-lane road with an entrance ramp (FIG. 3), a four-way-stop intersection with intersecting two-lane roads oriented at an angle (not limited to a right-angle), a residential roadway lined with driveways to individual residences, and a round-about type intersection with multiple roadways joined to the round-about at a variety of angles.

[0017] The traffic-scenario detector 14 may be or may include, but is not limited to, a camera, a radar-unit, a lidar-unit, or any combination thereof that could be useful to characterize or determine the present-scenario 16 of the host-vehicle 12, where the present-scenario 16 is determined based on a signal from the traffic-scenario detector 14. In addition, or as an alternative, the traffic-scenario detector 14 may include a location-indicator 20 that, for example, determines the global-coordinates of the host-vehicle 12 so the system 10 can determine the present-scenario 16 by consulting a digital-map 22 that indicates, for example, the number of lanes of the roadway 18, presence of an entrance or exit ramp, intersection controls (e.g. traffic-signal or stop-sign), and the like. That is, the traffic-scenario detector 14 may include a location-indicator 20, and the present-scenario 16 may be determined based on a map-location 40 of the host-vehicle 12 on the digital-map 22 as indicated by the location-indicator 20.

[0018] The system 10 also includes an object-detection device 24 used to detect an object 26 proximate to the host-vehicle 12. The object-detection device 24 may be or may include, but is not limited to, a camera, radar-unit, lidar-unit, or any combination thereof that could be useful to identify or classify the object 26. The object 26 may be, but is not limited to, the roadway 18, features that define boundaries of the roadway 18, an other-vehicle 28, a fixed-object 30 such as a traffic-barrier, building, sign, tree, or any other instance of the object 26 that could be the fixed-object 30.

[0019] The system 10 also includes a controller 32 in communication with the traffic-scenario detector 14 and the object-detection device 24. The controller 32 may include a processor 34 such as a microprocessor or other control circuitry such as analog and/or digital control circuitry including an application specific integrated circuit (ASIC) for processing data, as should be evident to those in the art. The controller 32 may include memory 42, including non-volatile memory, such as electrically erasable programmable read-only memory (EEPROM) for storing one or more routines, thresholds, and captured data. The one or more instances of possible-scenarios 44 and/or optimized-algorithms 46 that may be used the processor to perform steps to determine a preferred-algorithm 36 used to identify or determine an object-identification 38 of the object 26, where the preferred-algorithm 36 is determined based on the present-scenario 16, as will be described in more detail below. Accordingly, the controller 32 may be configured to select the present-scenario 16 from a plurality of possible-scenarios 44 stored in the memory 42.

[0020] FIG. 2 illustrates a non-limiting example of the present-scenario 16 that may be characterized as a multilane-expressway where the other-vehicle is traveling in an adjacent-lane 50 that is adjacent to or next to a travel-lane 52 of the host-vehicle 12. The controller 32 may be tracking the other-vehicle 28 using the object-detection device 24. Information from the traffic-scenario detector 14 such as data from the digital-map 22 and/or images from the camera indicate that the present-scenario 16 is a multilane-expressway. The preferred-algorithm 36 selected from the optimized-algorithms 46 is optimized for tracking the other-vehicle 28 along an expected-direction 54 that is parallel to and offset from a travel-direction 56 of the host-vehicle 12. As such, the controller 32 is configured to further determine or select the preferred-algorithm 36 based on the expected-direction 54 of motion of the object 26 relative to the host-vehicle 12 indicated by the present-scenario 16.

[0021] The preferred-algorithm 36 may be selected for processing signals from the camera or radar-unit because the perspective the other-vehicle 28 is quartering away from the host-vehicle 12. That is, an image of or radar-reflection from the other-vehicle will likely include data-points that correspond to the tail-end and left-side of the other-vehicle. By way of further example, the processor 34 may attempt to match the present-image from the camera to one of a plurality of previously stored images, or match the radar-reflection to a predetermined reflection-pattern.

[0022] The preferred-algorithm 36 may also be selected or optimized to detect lateral motion of the other-vehicle 28 which would occur if the other-vehicle 28 executed a lane-change 58 i.e. moves to a position in front of the host-vehicle 12 or ‘cuts-in’ to the travel-lane 52. Because an optimized algorithm was selected to monitor for lateral motion, the system is able to identify and track the cutting-in by other-vehicle 28 faster and more reliably. The preferred-algorithm may also selected by using the digital-map 22 since the relative location of the adjacent-lane 50 is known. The system then tracks the closest of other-vehicles leading the host-vehicle 12 in neighboring lanes and computes their lateral velocity. In response to detecting that the other-vehicle 28 is cutting in, the host-vehicle 12 may begin to perform distance keeping relative to the other-vehicle after the cutting-in event.

[0023] FIG. 3 illustrates a non-limiting example of the present-scenario 16 that may be characterized as a two-lane road 60 with an entrance-ramp 62 that will merge the other-vehicle into the travel-lane 52. The presence of the entrance-ramp 62 may be determined from the digital-map 22 based on information from the location-indicator 20. The preferred-algorithm 36 may be selected from the optimized-algorithms 46 as one that is optimized to track a vehicle traveling beside but not necessarily parallel to the host-vehicle 12, and one that can readily determine the relative-speed of the other-vehicle 28 with respect to the host-vehicle 12. Based on the relative-speed and acceleration of the other-vehicle 28 with respect to the host-vehicle 12, the system 10 computes a time-to-arrival to a merging-point 64 for the host-vehicle 12 and the other-vehicle 28, and determines if a conflict is likely. The host-vehicle 12 may elect to slow-down or speed-up depending on the location of the merging-point 64 and/or relative times-to-arrival.

[0024] Accordingly, a scenario aware perception system (the system 10), a controller 32 for the system 10, and a method of operating the system 10 is provided. The preferred-algorithm used to process signals from the object-detection device 24 is selected based on the present-scenario 16 being experienced by the host-vehicle 12. By selecting an algorithm that has been optimized for the present-scenario, the reliability of tracking the object 26, e.g. the other-vehicle 28, is improved.

[0025] While this invention has been described in terms of the preferred embodiments thereof, it is not intended to be so limited, but rather only to the extent set forth in the claims that follow.