PROCEDURES FOR RISK ASSESSMENT IN THE CASE OF TRAFFIC SIGNAL VIOLATIONS

Abstract

System and procedure for risk assessment in the event of a traffic signal violation for a vehicle. The system comprises a sensor, such as a front camera module (FCM), configured to capture an image in a field of view in front of the vehicle; a vehicle communication system configured for vehicle-to-infrastructure (V2I) and / or telematics communication with an infrastructure unit, the infrastructure unit being capable of providing intersection information; and a traffic signal violation risk assessment module (TSVRA module) that communicates with the FCM and the vehicle communication system.The TSVR module analyzes the captured image to determine an intersection, a traffic light at the intersection, and the current state of the traffic light (TLCS, Traffic Light Current State); it gathers intersection information from the infrastructure unit and determines a risk of a traffic light signal violation based on the information and performs a physical action to mitigate the risk.

Description

INTRODUCTION

[0001] The present disclosure relates to the assessment of collision risks for vehicles in motion, and more particularly to a system and method for risk assessment and mitigation of traffic light signal violations.

[0002] The document US 2021 / 0 065 551 A1 discloses a method for risk assessment in the event of a traffic light signal violation according to the preamble of claim 1.

[0003] Object detection systems and collision avoidance systems are widely used in modern vehicles. Such systems can perform a risk assessment and warn the driver of a moving object in the path of a host vehicle. The warning may include a visual indicator on a vehicle's dashboard, an audible warning such as a chime, and / or a haptic warning via a feedback device installed on a user interface such as the steering wheel. Object detection systems can also provide input to active vehicle systems, such as adaptive cruise control systems, that regulate the vehicle's speed to maintain an appropriate longitudinal following distance from a remote vehicle ahead. Collision avoidance systems can provide both warnings and automatic braking to avoid an impending collision with an object in the vehicle's path.

[0004] Object detection and collision avoidance systems fulfill their purpose of detecting objects, assessing collision risk, and mitigating potential collisions with objects in the path of vehicles. However, there is a need for a system and / or method for traffic signal detection, assessing the risk of a traffic signal violation by a host vehicle and other vehicles approaching the traffic signal, and mitigating such risks. SUMMARY

[0005] According to several aspects, a method for risk assessment in the event of a traffic signal violation is provided. The method includes approaching a host vehicle to an intersection, determining whether the intersection has a traffic signal, and obtaining intersection information. The intersection information includes traffic signal data including the current state of the traffic signal (TSCS) and the time to next state (TTNS).The method further includes determining, by the host vehicle, a time of arrival at the intersection (TAI), determining, by the host vehicle, a risk level of a traffic signal violation based on TSCS, TTNS, and TA1, and implementing, by the host vehicle, a first remedial action in response to the risk level determined as low risk, and a second remedial action in response to the risk level determined as high risk. The first remedial action includes a unicast alert, and the second remedial action includes a geocast alert. The second remedial action further includes a vehicle control module assuming partial control of the vehicle. The partial control includes emergency braking, steering control, and / or preconditioning of the vehicle, e.g., seatbelt pretensioning.

[0006] As another aspect of the present disclosure, the TSCS includes either a GO state or a STOP state. The method further includes determining that the TSCS is the STOP state and TAI is less than TTNS, or that the TSCS is the GO state and TAI is greater than TTNS, before determining a risk level of a traffic signal violation by the vehicle.

[0007] In another aspect of the present disclosure, the TSCS and TTNS data of the traffic light signal are wirelessly obtained from a remote server, a roadside unit (RSU) and / or a mobile edge computer (MEC) communicating with the traffic light signal.

[0008] In another aspect of the present disclosure, the host vehicle includes a forward-facing camera configured to capture images of dynamic traffic signal signals in front of the vehicle. The TSCS of the traffic signal is determined based on the captured images.

[0009] In another aspect of the present disclosure, the method further includes determining a current attention level of a driver of the host vehicle and determining an intensity of sunlight based on a horizontal direction of the sunlight and whether the sunlight is directly within a cone of vision of the driver. The host vehicle's determination of the risk level of a traffic signal violation is further based on the current attention level of the driver and the intensity of the sunlight.

[0010] In another aspect of the present disclosure, the intersection information further includes traffic information about remote vehicles approaching the intersection and the determination of how many times the host vehicle has previously passed through the intersection. The determination of a risk level of a traffic signal violation by the host vehicle is further based on the traffic information about vehicles approaching the intersection and the familiarity with the intersection based on how many times the host vehicle has previously passed through the intersection.

[0011] In another aspect of the present disclosure, determining the risk level of a traffic signal violation by the host vehicle comprises integrating intersection information, the current driver attention level, the intensity of sunlight, and intersection familiarity using a lookup table or fuzzy logic approach.

[0012] In another aspect of the present disclosure, the intersection information further includes traffic information about a remote vehicle approaching the intersection. The method further includes determining, by the remote vehicle, a time of arrival at the intersection (TAI) and determining, by the remote vehicle, a risk level of a traffic signal violation based on the TSCS, TTNS, and TAI.

[0013] According to several aspects, a traffic signal violation risk assessment system for a vehicle is provided. The system includes a vehicle sensor, such as a front camera module (FCM), configured to capture image data in a field of view in front of the vehicle; a vehicle communication system configured for vehicle-to-infrastructure (V2I) and / or telematic communication with an infrastructure entity, wherein the infrastructure entity is capable of providing intersection information; and a traffic signal violation risk assessment module (TSVRA) in communication with the FCM and the vehicle communication system.The TSVRA module is configured to: analyze the captured image data to determine an intersection, a traffic light at the intersection, and a traffic light current state (TLCS); collect intersection information from the infrastructure unit, wherein the intersection information includes the time to next state (TTN) of the traffic light, traffic light current state at the intersection; determine a risk of a traffic light signal violation based on TLCS and TTN; and perform a physical action to mitigate the risk of a traffic light signal violation.

[0014] In an additional aspect of the present disclosure, when determining the risk of a traffic signal violation, a distinction is made between low and high risk. The physical action includes unicasting an alert in response to a specific low risk and geocasting an alert in response to a specific high risk. Geocasting includes changing the current state of the traffic signal and / or activating a strobe light at the intersection.

[0015] In another aspect of the present disclosure, the TSVRA module is further configured to determine the vehicle's time of arrival at the intersection (TAI). The TLCS includes either a GO or a STOP state. The TLCS is the STOP state and the TAI is less than the TTNS, OR the TSCS is the GO state and the TAI is greater than the TTNS, before determining the risk of the traffic signal violation.

[0016] In another aspect of the present disclosure, the system further includes an ambient light sensor module configured to determine sunlight intensity and a driver monitoring system configured to determine the current attention level of a driver of the host vehicle. The TSVRA module is further configured to determine the risk of a traffic signal violation based on TLCS, TTN, sunlight intensity, and the current driver attention level.

[0017] In another aspect of the present disclosure, the TSVRA module is further configured to use a lookup table or fuzzy logic to combine intersection traffic information, sunlight intensity, the driver's current attention level, and intersection familiarity to determine the risk of a traffic signal violation by at least one of the host vehicles.

[0018] According to several aspects, a traffic signal violation risk assessment (TSVRA) module is provided. The TSVRA module comprises a processor; a computer-readable medium with instructions that cause the processor to: analyze an image of a front view of a vehicle to identify an intersection; determine that the intersection includes a traffic signal; communicate with infrastructure entities to determine a current state of the traffic signal, a time until the next change in state of the traffic signal, and vehicular traffic approaching the intersection; determine a risk level of a traffic signal violation by a vehicle based on the current state of the traffic signal, the time until the next change in state of the traffic signal, and vehicular traffic approaching the intersection, wherein the risk level includes one of low risk and high risk;and unicasting an alert in response to a specific low risk and geocasting a multimodal proactive alert in response to a specific high risk, together with instructing a vehicle control unit to assume at least partial control of the vehicle in response to the specified high risk;

[0019] Further areas of applicability will become apparent from the following description. It should be understood that the description and specific examples are for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are for illustrative purposes only. Fig. 1 illustrates a plan view of a road intersection with a dynamic traffic signal according to an exemplary embodiment; Fig. Figure 2 is a high-level functional flow diagram for a method for assessing the risks of a traffic light signal violation by a vehicle approaching the intersection of Fig. 1, according to an exemplary embodiment; Fig. 3 is a functional diagram of a connected host vehicle according to an exemplary embodiment; Fig. 4 is a functional block diagram of a traffic signal violation risk assessment (TSVRA) module according to an exemplary embodiment; Fig. 5 is a flowchart of a traffic signal violation risk assessment method according to an exemplary embodiment; Fig. 6 is a flowchart of a method for determining initiation of a traffic signal violation risk assessment procedure according to another exemplary embodiment; and Fig. 7 is a fusion logic lookup table for evaluating the risk level of a traffic signal violation according to an exemplary embodiment. DETAILED DESCRIPTION

[0021] The following description is merely exemplary. The illustrated embodiments are disclosed with reference to the drawings, wherein like numerals designate corresponding parts throughout the various drawings. The figures are not necessarily to scale, and some features may be greatly exaggerated or reduced in size to show details of particular features. The specific structural and functional details disclosed are not to be considered limiting, but rather as a representative basis for showing those skilled in the art how to practice the disclosed concepts.

[0022] As used herein, the term module, control module, or control device refers to any hardware, software, firmware, electronic control component, processing logic, and / or processor device, individually or in any combination, including, but not limited to: an application-specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or as a group), and memory executing one or more software or firmware programs, a combinational logic circuit, and / or other suitable components that provide the described functionality.

[0023] Embodiments of the present disclosure may be described herein in terms of functional and / or logical block components and various processing steps. It should be appreciated that such block components may be implemented by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of the present disclosure may utilize various integrated circuit components, e.g., memory elements, digital signal processing elements, logic elements, lookup tables, or the like, which may perform a variety of functions under the control of one or more microprocessors or other control devices.Furthermore, those skilled in the art will appreciate that embodiments of the present disclosure may be used in connection with any number of systems and that the systems described herein are merely exemplary embodiments of the present disclosure.

[0024] The connecting lines depicted in the various figures included herein are intended to represent exemplary functional relationships and / or physical connections between the various elements. Conventional techniques may be used for signal processing, data transmission, signaling, control, and other functional aspects of the systems (and the individual operating components of the systems) and may not be described in detail herein. It should be understood that many alternative or additional functional relationships or physical connections may be present in an embodiment of the present disclosure.

[0025] Fig. 1 illustrates a non-limiting example of a plan view of an intersection 100 having a dynamic traffic signal 102 configured to direct vehicular traffic flow through the intersection 100. The intersection 100 is defined by a first lane 104 intersecting a second lane 106. Both the first lane 104 and the second lane 106 are configured for bidirectional traffic. The first lane 104 includes a pavement marking 108, depicted as a white solid line 108, perpendicular to the first lane 104 and located in front of the intersection 100 in the direction of travel toward the intersection 100. The white solid line 108 is referred to as a stop line 108 and indicates a stopping point for a vehicle responding to a stop signal or command issued by the traffic signal 102.

[0026] Three predetermined decision zones are shown in the direction of travel toward the intersection. The predetermined decision zones are referred to as Zone A, Zone B, and Zone C. Zone A is farthest from stop line 108 compared to Zones B and C. Zone B is further from stop line 108 than Zone C. Zone C is immediately adjacent to stop line 108. Zone A is a probable stop zone, Zone B is the decision or dilemma zone, and Zone C is the probable GO zone. The dilemma zone (Zone B) is an area of ​​an intersection approach where a driver is faced with a conditional traffic signal, such as a yellow signal display. In response to the conditional traffic signal, the driver must decide whether to continue through intersection 100 or to stop before intersection 100.The dilemma zone allows an estimated time of between 2.5 and 5.5 seconds to cause the vehicle to stop before reaching the stop line 108 in response to a decision to stop before intersection 100.

[0027] The dynamic traffic signal 102 is located at the intersection 100 and is visible to vehicles approaching the intersection 100 on the first lane 104 and the second lane 106. The traffic signal 102 is capable of alternating between visual indications, also referred to as phases or states, to direct the flow of vehicle traffic through the intersection 100. Common visual indications may include words, symbols, and / or colors. In one non-limiting example, the color indications include a green state, a yellow state, and a red state to instruct vehicles to continue through the intersection 100, to prepare to stop before the intersection 100, and to stop at the intersection 100, respectively.

[0028] A traffic camera 110 is shown located at intersection 100 near traffic signal 102 and configured to monitor vehicle traffic approaching and passing through the intersection. Traffic camera 110 may be used to monitor weather and road conditions near intersection 100. Traffic camera 110 may be pointed in the direction of approaching traffic and synchronized with the state change of the traffic signal to detect vehicles at risk of violating the traffic signal. Additional cameras 112 may be mounted around intersection 102 to monitor vehicle and pedestrian traffic crossing the intersection.

[0029] A pedestrian 114 with a personal communications device 116, such as a cellular phone 116, approaches a crosswalk 118 extending across the first lane 104. At one end of the crosswalk 118, a pedestrian signal 120 is shown, configured to direct pedestrian traffic across the first lane 104. The pedestrian signal 120 may include a visual and / or audible indication to instruct pedestrians to cross or not cross the lane.

[0030] A host-connected vehicle 200, or host vehicle 200, is shown approaching the intersection on the first lane. A remote-connected vehicle 124 is shown approaching intersection 100 on the second lane 106. A non-connected remote vehicle 125 is shown approaching intersection 100 in the opposite direction of the remote-connected vehicle 124. The host vehicle 200 is in communication with a traffic signal violation risk assessment (TSVRA) system 300, which may be onboard the host vehicle 200, off-board the host vehicle 200, or partially onboard and off-board the host vehicle 200. The TSVRA system may be configured for vehicle-to-everything (V2X) communication.

[0031] At least one Roadside Unit (RSU) 126 and / or a Mobile Edge Computer (MEC) 130 are located near the intersection 100. The RSU 126 and the MEC 130 are configured to collect intersection data and transmit it to traffic control devices such as the traffic lights 102, 120, as well as to central traffic control centers. The RSU 126 and the MEC 130 are configured to communicate with remote servers 128, e.g., servers in the cloud or in back offices, and / or with the cellular infrastructure 132 to upload or retrieve data relevant to the safe operation of the intersection. The RSU, the MEC 130, the remote servers 128, and / or the cellular infrastructure 132 can wirelessly communicate the intersection data to the connected vehicles traveling within a predetermined boundary 131 around the intersection 100.Such intersection data may include, but is not limited to, the location of the intersection, the state of the traffic light, the time until the next traffic light state, information about approaching vehicles, road conditions, weather conditions, sun glare, etc. The RSU 126 and the MEC 130 may be configured to communicate with connected vehicles approaching the intersection and with personal devices either directly via infrastructure-to-vehicle (I2V) communication, wireless telematics services 132, and / or the Internet.

[0032] Fig. Figure 2 is a high-level functional flow diagram for a method for assessing a risk of traffic light signal violation by a vehicle approaching the intersection of Fig. 1. A traffic signal violation is defined as a vehicle proceeding through an intersection despite the traffic signal indicating a stop command for that vehicle. A non-limiting example of a traffic signal violation may occur when a vehicle fails to stop at a red light. The vehicle may be the host vehicle 200, the networked remote vehicle 124, and the non-networked remote vehicle 125. The TSVRA system 300 collects information from the host vehicle 200, with networked remote vehicles 124 using V2V communication and infrastructure-based sources using I2V communication. The infrastructure-based sources include, but are not limited to, the traffic signal 102, cameras 110, 112, RSU 126 and MEC 130, and remote server 128.

[0033] The information collected by the host vehicle 200 includes, among other things, data captured by a front-facing camera module (FCM), a driver monitoring system (DMS), and an ambient light sensor module. The FCM is configured to detect the status of traffic lights, remote vehicles, and vulnerable road users (VRUs) such as pedestrians and cyclists. The DMS monitors the driver's attention level while operating the vehicle. The ambient light sensor module can be used to determine the intensity of sunlight based on the horizontal direction of the sunlight and whether the sunlight is directly in the driver's line of sight.

[0034] Information collected from infrastructure-based sources includes, but is not limited to, traffic signal status and time to the next status, traffic camera information, historical intersection safety records, and historical host vehicle map data for assessing the risk of a red light violation. The host vehicle's historical map data may indicate whether the driver of host vehicle 200 is familiar with intersection 100.

[0035] The information collected by the connected remote vehicle 124 as it approaches the intersection includes, among other things, the traffic light signal observed by the connected remote vehicle, the time of arrival at the intersection, and observations of the behavior of other vehicles and / or pedestrians at or near the intersection.

[0036] The collected information is analyzed to determine the degree of risk that a vehicle, including the host vehicle 200, the connected remote vehicle 124, and the non-connected remote vehicle 125, could violate the traffic signal 102. An information fusion model such as a lookup table, a fuzzy logic-based approach, and / or a weighting scheme that combines the three criteria using weights inversely proportional to the uncertainty of each element, quantified by its variance, or a Dempster combination rule that considers the degree of credibility of each criterion to determine a risk level ranging from no risk to low risk to high risk.

[0037] As a non-limiting example, no action is taken if the risk level is determined to be an unknown risk or an unknown or zero risk; a unicast alert may be issued to a human-machine interface 250 within the host vehicle 200 if the risk level is determined to be a low risk level; or a geocast alert may be sent to personal communication devices 116 of VRUs and networked remote vehicles 124 if the risk level is determined to be a high risk level. The geocast alert may also be sent to the traffic signal 102 and / or the pedestrian signal 120 to change the current state of the traffic signal and / or activate a strobe light to mitigate a high risk level. Furthermore, if the host vehicle 200 is equipped with an autonomous driving function, such asbraking or steering, the TSVRA System 300 may also activate the ADS, and / or the ADS may mitigate the potential for a collision event.

[0038] Fig. 3 is a functional diagram of a networked host vehicle 200 in communication with a traffic signal violation risk assessment (TSVRA) system 300. Although the TSVRA system 300 is shown external to the host vehicle 200, it is contemplated that some or all of the components defining the TSVRA system 300 may be part of the host vehicle 200. In one non-limiting example, the TSVRA system 300 may be a cloud-based system that wirelessly communicates with the host vehicle 200. In another non-limiting example, the TSVRA system 300 may be a vehicle system that communicates with the various other vehicle systems onboard the vehicle.

[0039] The vehicle 200 generally includes a body 206, front wheels 208, and rear wheels 210. The body 206 substantially encloses the systems and components of the vehicle 200. The front wheels 208 and the rear wheels 210 are each rotatably connected to the body 206 near a corresponding corner of the body 206. Although the host vehicle 200 is depicted as a sedan, it is contemplated that the host vehicle 200 may be another type of vehicle, such as a pickup truck, a coupe, a sport utility vehicle (SUV), a motor home, motorcycles, and non-motorized road vehicles, such as bicycles, that have wireless connectivity infrastructure and / or the Internet and utilize personal communication devices 116, or the like.

[0040] Regardless of the type of vehicle, the host vehicle 200 is capable of communicating with other connected vehicles, connected roadside units (RSU), or cloud computing services via wireless cellular communication, Dedicated Short-Range Communication (DSRC), Cellular C-V2X (C-V2X) communication. The host vehicle 200 may be an intelligent vehicle equipped with an advanced driver-assistance system (ADAS) configured to enhance the safe operation of the vehicle and / or equipped with an automated driving system (ADS) capable of operating from Level 0 (no driving automation) to Level 5 (full driving automation) according to the SAE J3016 levels of driving automation.

[0041] As illustrated, the vehicle 200 generally includes a drive system 220, a transmission system 222, a steering system 224, a braking system 226, a vehicle communication system 230, a driver monitoring system (DMS) 232, an ADAS 238 and / or ADS 239, a plurality of sensors 240, and actuators 242. The vehicle 200 further includes one or more vehicle control units 243, also called controllers 243, that communicate with one or more vehicle systems, sensors 240, 234, 236, and actuators 242. The controller 243 is also in communication with the TSVRA system 300 to assess the risk of traffic signal violations and with the ADAS 238 and / or ADS 239 to mitigate such risks.

[0042] The plurality of sensors 240 collects information and generates sensor data indicative of the collected information. As non-limiting examples, the sensors 240 may include, but are not limited to, Global Navigation Satellite System (GNSS) transceivers or receivers, yaw rate sensors, speed sensors, lidars, radars, ultrasonic sensors, and external and in-cabin cameras. The GNSS transceivers or receivers are configured to detect the location of the host vehicle 200. The speed sensors are configured to detect the speed of the host vehicle 200. The yaw rate sensors are configured to determine the heading of the host vehicle 200.

[0043] Sensors 240 further include a front camera module (FCM) 234 and an ambient light sensor module 236. The field of view of FCM 234 is large enough to capture images in front of vehicle 200. FCM 234 is configured to detect dynamic traffic signal signals, such as traffic lights, and detect the change in state of the dynamic traffic signal signals. It should be noted that vehicle sensors 240 may include a LIDAR (Light Detection and Ranging) unit, a thermal scanner, and other external data sensors that may be configured to detect dynamic traffic signal signals and detect the change in state of the dynamic traffic signal signals. Ambient light sensor module 236 is configured to detect the intensity of sunlight based on the horizontal direction of the sunlight and whether the sunlight is directly within the driver's line of sight.

[0044] The vehicle control unit 243 includes at least one vehicle processor 244 and a non-transferable computer-readable vehicle storage device or storage medium 246. The vehicle processor 244 may be a custom or off-the-shelf processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor among multiple processors networked to the vehicle control unit 243, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, a combination thereof, or generally a device for executing instructions. The computer-readable vehicle storage device or storage medium 246 may include, as non-limiting examples, volatile and non-volatile memory in read-only memory (ROM), random access memory (RAM), and keep-alive memory (KAM).KAM is a persistent or non-volatile memory that can be used to store various operating variables while the vehicle processor 244 is powered off. The computer-readable vehicle storage device or vehicle storage medium 246 of the vehicle control module 243 can be implemented using a variety of storage devices such as PROMs (Programmable Read-Only Memory), EPROMs (Electrical PROMs), EEPROMs (Electrically Erasable PROMs), flash memory, or other electrical, magnetic, optical, or combination storage devices capable of storing data, some of which represent executable instructions used by the vehicle control module 243 in controlling the host vehicle 200. The non-transferable computer-readable vehicle storage medium or vehicle storage medium 246 can store map data and / or sensor data received from one of the sensors 240.The sensor data may include location data received from the GNSS transceiver. The map data includes a navigation map.

[0045] The vehicle communication system 230 may include one or more communication transceivers 247. Each of the communication transceivers 247 is configured to wirelessly communicate information to and from other remote entities, such as networked remote vehicles 124, RSU 126, MEC 130, infrastructure (through "V2I" communication), remote servers 128 in the cloud or in a back office or remote call center, and / or personal electronic devices 116, such as a smartphone. The communication transceiver 247 may be configured to communicate via a wireless local area network (WLAN) conforming to the IEEE 802.11 standard or via cellular data communications. However, additional or alternative communication methods, such as Cellular V2X (C-V2X) or Dedicated Short-Range Communications (DSRC), are also contemplated within the scope of the present disclosure.DSRC channels refer to short- to medium-range, one-way or two-way wireless communication channels specifically designed for use in vehicles, along with a corresponding set of protocols and standards. Accordingly, the communication transceiver 247 may include one or more antennas for receiving and / or transmitting signals, such as Cooperative Sensing Messages (CSMs). The communication transceiver 247 may be considered data sources.

[0046] With reference to Fig. 4, the TSVRA system 300 includes one or more TSVRA controllers 334. The TSVRA controller 334 is configured to consider the potential causes of traffic signal violations such as sun glare (e.g., glare from the sun preventing the driver from seeing the red light), distraction (e.g., distracted or inattentive driver), driving assessment / recording (e.g., aggressive driver), uncertainty (e.g., drivers unfamiliar with the intersection, and inexperienced drivers).

[0047] The TSVRA controller 334 includes at least one system processor 344 and a non-transferable computer-readable system storage device or medium 346. The system processor 344 may be a custom or off-the-shelf processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor among multiple processors networked to the TSVRA controller 334, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, a combination thereof, or generally a device for executing instructions. The computer-readable system storage device or medium 346 may include volatile and non-volatile memory, for example, read-only memory (ROM), random access memory (RAM), and keep-alive memory (KAM).The computer-readable system storage device or medium of the TSVRA controller 334 may be implemented using a variety of storage devices such as PROMs (programmable read-only memory), EPROMs (electrical PROMs), EEPROMs (electrically erasable PROMs), flash memory, or other electrical, magnetic, optical, or combination storage devices capable of storing data, some of which represent executable instructions.

[0048] The non-transitory, computer-readable system storage device or medium 346 and / or the non-transitory, computer-readable vehicle storage device or medium 346 contain machine-readable instructions that, when executed by the one or more system processors 344 and / or vehicle processors 244, cause the system processors 344 and / or vehicle processors 244 to perform the methods described below. The TSVRA controller 334 is in communication with the vehicle controller 243, the vehicle systems, and sensors, including, but not limited to, the communication systems 230, DMS 232, FCM 234, and the ambient light sensor module 236. The TSVRA controller 334 is also in communication with infrastructure units, including, but not limited to, the RSU 126, the remote server 128, the MEC 130, and the personal communication devices 116.

[0049] The method 400 begins at block 402 when the host vehicle 200 is traveling on a roadway. At block 404, the TSVRA controller 334 monitors the location of the host vehicle 200 and determines whether the host vehicle 200 is approaching an intersection with a dynamic traffic signal. Continuing with block 406, if the host vehicle 200 is not approaching an intersection with a dynamic traffic signal, the method 400 returns to block 402. Returning to block 406, if the host vehicle is approaching an intersection with a dynamic traffic signal, the method continues to block 408.

[0050] At block 408, the TSVRA controller 334 determines whether the host vehicle 200 is within a predetermined decision zone relative to a stop line at the intersection. If the host vehicle 200 is not within the predetermined decision zone, the method 400 returns to block 402. Returning to block 408, if the vehicle is within the predetermined decision zone, the method 400 continues to block 410.

[0051] At block 410, the TSVRA controller 334 collects and analyzes information from the host vehicle 200, information about the current state of the traffic signal and the time to the next state, which it receives from the RSU 126, the MEC 130, and / or the remote servers 128 in the cloud or back offices, as well as traffic information from networked remote vehicles 124 near the intersection. The TSVRA controller 334 may also analyze the information received from the traffic cameras 110, 112 to determine whether other remote vehicles 125, whether networked or not, are approaching the intersection.The host vehicle 200 information includes, but is not limited to, the current traffic signal state from the FCM 234, the driver distraction state from the DMS 232, the level of sun glare from the ambient light sensor module 236, familiarity with the intersection from the historical map data, and the number of times the host vehicle 200 has traveled through the intersection, as well as other information such as the driver's historical driving rating and an intersection rating based on crash records that are relevant as an indication of a potential risk of a traffic signal violation at a particular intersection.

[0052] Continuing with block 410, an information fusion approach may be used to combine the relevant information (i.e., detection of the current signal state, signal phase and timing of the traffic light signal, alerts from infrastructure or connected vehicles, driver distraction state, sun glare, and unknown intersections) to assess the risk level of a traffic light signal violation. This information fusion module may be a simple lookup table, a fuzzy logic-based approach, or a weighting scheme that combines the three criteria using weights inversely proportional to the uncertainty of each element, quantified by its variance, or a Dempster combination rule that considers the degree of credibility of each criterion.

[0053] As a non-limiting example, a fuzzy logic-based TSLVR estimation, also referred to as red light violation (RLV) risk, can be used to estimate whether the risk of a traffic signal violation is unknown, zero, low, or high. Indicators such as FCM-based red light detection, I2V-based signal phase and timing, V2V-based warning, DMS-based distracted driver detection, sun glare detection, and unknown intersection detection are subjected to a fuzzification process. The fuzzification of the six indicators is input into an inference engine that operates based on predetermined rules to estimate the risk level of an RLV. Input set: Starting sentence: M={Red light violation (RLV) risk}={m}

[0054] The total number of possible rules to which the fuzzy logic-based strategy is subject is given as I = k1k2...k n , where: n is the number of non-interactive inputs; and k i is the number of partitions of the input universe of discourse i, i ∈ {1, 2, ..., n}.

[0055] For 6 indicators with 4 membership functions such as “no, unknown, low, high”, the number of possible rules is 4 6= 4096. In practice, the actual number of rules required for the fuzzy inference rule base is significantly less than I due to the interpolative reasoning capability of the fuzzy model and the overlap of the fuzzy functions of the partitions. The strategy can be optimized by adding or modifying rules and by adjusting specified boundaries. These rules can be created manually by conducting some experiments to understand the relationship between the context information and the signal conditioning mode. The rules can also be learned automatically from the sample data using a learning algorithm such as neural networks or decision trees. Fusion neuro-fuzzy systems such as the Dynamic Evolving Neural-Fuzzy Inference System (DENFIS) can also be used.

[0056] The estimated RLV risk can be considered as an explicit context of interest represented as a linguistic variable; then its term set T(b) could be: T(b) = {No, Unknown, Low, High}. Each of the terms in T(b) is characterized by the fuzzy set in a discourse universe U = [-3, 3].

[0057] In a non-limiting example: If I FCM = 70% high AND I I2V = 85% to the power of AND I V2V = 90% unknown AND I DMS = 83% Yes AND I Sonnenblendung = 30% Yes AND I Unvertrautheit = 15% Yes, then RLV risk = 90% high.

[0058] In another non-limiting example: If I FCM = 82% No AND I I2V = 75% No AND I V2V = 84% No AND I DMS = 87% No AND I Sonnenblendung = 75% No AND I Unvertrautheit = 92% Yes, then RLV risk = 85% No risk.

[0059] The six indicators can also be combined in other ways, for example using a lookup table, as in Fig. 7, or a weighted scheme that combines the three criteria using weights inversely proportional to the uncertainty of each element, which is qualified by its variance, or using Dempster's combination rule, which takes into account the degree of credibility of each criterion.

[0060] Continuing with block 412, a risk mitigation measure can be activated based on the determined risk level. No action is taken if the risk level is determined to be zero risk. A proactive warning can be sent via geocast or unicast to surrounding vehicles via V2V, to infrastructure via V2I, or to vulnerable road users (VRUs) such as pedestrians / cyclists via cellular-based V2VRU communication via a safety app to which the VRU may be subscribed, such as the OnStar Guardian app.

[0061] As one non-limiting example, a unimodal proactive alert may be generated to warn the driver of the host vehicle 200 when the risk level is determined to be low. Proactive alerts may be a series of messages that can be preprogrammed to communicate to connected vehicles via V2V communication and to VRUs via personal communication devices 116, and / or audible messages sent via the infrastructure within the boundary 131 of the intersection 100. The unimodal alert may be an audible alert, a visual alert on an HMI device, or a tactile alert. A multimodal proactive alert may be generated to warn connected vehicles 124 and VRUs within a predetermined perimeter 131 bounding the intersection 102 when the risk level is determined to be high.The proactive alert may be unicast to a specific connected vehicle or vulnerable user (VRU) within the perimeter, or geocast to multiple connected vehicles and / or VRUs via the host vehicle's V2V communication system directly, RSU / MEC, and / or the cloud using V2V, V2I, and / or telematics communications. If the host vehicle 200 is equipped with an ADAS and / or ADS, the vehicle control module may initiate partial or full control of the host vehicle, e.g., emergency braking or steering, as well as vehicle preconditioning, such as seatbelt pretensioning, to mitigate a high risk determination. The method ends at block 414.

[0062] Fig.6 shows a flowchart that considers the host vehicle's time of arrival (TAI) at the stop line 108 at the intersection 100, the current traffic light state (TLCS) or the current traffic light signal state (TSCS), and the time to next state (TTNS) when assessing risk for a traffic light signal violation. The method 500 begins at block 502 when the host vehicle 200 is traveling on a road. Continuing with block 504, the host vehicle 200 retrieves map information from a remote server 128 located on the cloud server and / or in a back office. The map information includes the location of an upcoming intersection and whether the intersection has a traffic signal, such as a traffic light.

[0063] Continuing with block 506, if the upcoming intersection does not have a traffic light, the method 500 returns to block 502. Returning to block 506, if the intersection has a traffic light and the host vehicle 200 is within a predetermined distance of the intersection, the method 500 continues with block 508.

[0064] At block 508, the host vehicle 200 obtains information from the FCM 234 to determine whether the current state of the traffic light signal is red, yellow, or green. The host vehicle 200 attempts to obtain information from a remote server or RSU to confirm the current state of the traffic light (TLCS) and the time to next state (TTNS). The TSVRA controller 334 calculates the TAI based on the current speed of the host vehicle 200.

[0065] Continuing with block 510, if the TAI is greater than a predetermined TAI threshold, the method 500 returns to block 502. A non-limiting example of a TAI threshold may be approximately 5.5 seconds. Returning to block 510, if the TAI value is below the predetermined TAI threshold, the method 500 continues to block 512.

[0066] At block 512, if the TTNS is known, the method 500 proceeds to block 514. At block 514, the information from the FCM 234 is analyzed to determine whether the TLCS is in a no-travel command (RED) or in preparation for a stop command (YELLOW). If the TLCS: (i) is RED and the TAI to the stop line is less than TTNS, OR (ii) the TLCS is GREEN and the TAI is greater than TTNS, then the method 500 proceeds to block 518, where a traffic signal violation risk assessment is performed according to method 400. Otherwise, the method 500 returns to block 502.

[0067] Referring back to block 512, if the TTNS is unknown, the method 500 proceeds to block 516. At block 516, if the TLCS is neither RED nor YELLOW, the method 500 returns to block 502. Referring back to block 516, if the current state is RED or YELLOW, the method 500 proceeds to block 518. At block 518, a traffic signal violation risk assessment is performed according to method 400. If the traffic signal violation risk assessment is determined to be low risk, a unicast to alert the host vehicle may be a mitigating action.

[0068] Continuing with block 520, it is determined whether the TAI is less than the time required to stop at the stop line (braking threshold). If the TAI is not less than the braking threshold, the method returns to block 502. Referring back to block 520, if the TAS is less than the braking threshold, the method continues to block 522. At block 522, if a traffic signal violation risk assessment is determined to be a high risk level, a geocast alert is issued and a collision mitigation action, such as automatic braking, is performed if the host vehicle 200 is equipped with an ADAS 238 and / or ADS 239.

[0069] The present disclosure provides a wide range of signals and information to perform a risk assessment of a red light violation and take subsequent actions, which are communicated to vehicles and pedestrians via multiple channels: unicast, geocast, optical, etc.

Claims

[1] Procedure for risk assessment in the event of a traffic light signal violation, comprising: Approaching a host vehicle (200) to an intersection (100); Determining, by a module, that the intersection (100) has a traffic light signal (102); Obtaining intersection information, wherein the intersection information includes traffic signal data comprising a current state of the traffic signal (TSCS, Traffic Signal Current State) (102) and a time to next state (TTNS, Time To Next State); Determining the arrival time of the host vehicle (200) at the intersection (100) (TAI, Time of Arrival at the Intersection); Determining a risk level of a traffic light signal violation by the host vehicle (200) based on TSCS, TTNS, and TAI by the host vehicle (200), wherein the risk level of the traffic light signal violation includes one of the following: zero risk level, low risk level, and high risk level; and Implementing a first remedial action in response to the risk level determined as low risk and a second remedial action in response to the risk level determined as high risk, wherein the first remedial action and the second remedial action comprise one of a unicast alert or a geocast alert, characterized by , that the second remedial action further comprises assuming partial control of the host vehicle (200) by a vehicle control module, wherein the partial control comprises emergency braking, steering control, and / or preconditioning of the vehicle. [2] The method of claim 1, wherein the TSCS includes either a GO state or a STOP state; and further includes determining that the TSCS is the STOP state and TAI is less than TTNS, or that the TSCS is the GO state and TAI is greater than TTNS, before determining a risk level of a traffic signal violation by the host vehicle (200). [3] The method of claim 1, wherein the host vehicle (200) includes a forward-facing camera configured to capture images of dynamic traffic light signals (102) ahead of the host vehicle (200); and wherein the TSCS of the traffic light signal (102) is determined based on the captured images. [4] The method of claim 1, wherein the TSCS and TTNS of the traffic light signal (102) are wirelessly obtained from a remote server (128), a roadside unit (RSU) (126), and / or a mobile edge computer (MEC) (130) in communication with the traffic light signal (102). [5] The method of claim 1, further comprising: Determining a current attention state of a driver of the host vehicle (200) by a driver monitoring system (DMS) (232); and determining an intensity of the sunlight by an ambient light sensor module (236) based on a horizontal direction of the sunlight and whether the sunlight is directly in a driver's line of sight; and wherein determining the risk level of a traffic light signal violation by the host vehicle (200) is further based on the current driver attention level and the intensity of sunlight. [6] The method of claim 5, wherein the intersection information further includes traffic information about remote vehicles (124) approaching the intersection (100); and wherein determining a risk level of a traffic signal violation by the host vehicle (200) is further based on the traffic information about remote vehicles (124) approaching the intersection (100). [7] The method of claim 6, further comprising: Obtaining a location of the intersection (100); Determining an intersection rating and / or a number of trips the host vehicle (200) has made so far across the intersection (100); and Determining a familiarity with the intersection (100) based on the number of trips the host vehicle (200) has made through the intersection (100) so far. [8] The method of claim 7, wherein determining the risk level of a traffic signal violation by the host vehicle (200) comprises integrating the intersection information, the current driver attention level, the intensity of sunlight, and the familiarity with the intersection (100) using a lookup table or a fuzzy logic approach. [9] The method of claim 1, wherein the intersection information further includes traffic information about a remote vehicle (124) approaching the intersection (100); and further comprising: Determining a time of arrival at the intersection (TAI) (100) by the remote vehicle (124); and Determining a risk level of a traffic light signal violation by the remote vehicle (124) based on TSCS, TTNS and TAI by the remote vehicle (124).

Images