SYSTEMS AND METHODS FOR PREDICTING TRAFFIC SIGNAL PHASE AND TIMER

Abstract

A system for predicting a traffic signal phase and timing (SPaT prediction system) is disclosed. The system may include a transceiver configured to receive historical traffic signal information associated with a traffic light and real-time traffic signal information. The system may further include memory configured to store training data and a trained machine model. The trained machine model may be trained using the training data, which includes the historical traffic signal information. The system may also include a processor configured to execute instructions stored in the trained machine model to predict traffic signal information (e.g., SPaT) associated with a future traffic signal cycle based on the real-time traffic signal information.The processor can also output the traffic signal information associated with the future traffic signal cycle to a vehicle.

Description

REGIONThe present disclosure relates to systems and methods for predicting traffic signal phase and timing for a traffic light based on historical and real-time traffic signal information.GENERAL STATE OF THE ARTIt is known that vehicles consume a significant amount of fuel and / or battery charge on traffic lights when the traffic signal phase is red or yellow. A significant increase in vehicle operating efficiency may be achieved if the vehicles spend less time waiting for the traffic lights to make the traffic signal phase green, or if the vehicles cross the traffic lights when the traffic signal phase is green.To achieve greater vehicle operating efficiency, it is mandatory to accurately predict the signal phase and timing (SPaT) associated with the traffic lights. Conventional SPaT prediction systems are capable of predicting SPaT when the traffic lights are operated on a fixed schedule; however, the conventional systems have limited utility in predicting SPaT for traffic lights that are dynamically operated based on traffic conditions.Therefore, a system is required that accurately predicts SPaT for dynamically operated traffic lights.SUMMARYThe present disclosure describes a traffic signal phase and timing prediction (SPaT) system configured to predict traffic signal information, e.g., SPaT, associated with a traffic light for a future (e.g., next) traffic signal cycle. The system may predict the SPaT information for the future traffic signal cycle based on historical and real-time traffic signal information associated with the traffic light. The historical traffic signal information may be SPaT information associated with the traffic light for a predefined elapsed time period or a first predefined number of traffic signal cycles. The real-time traffic signal information may be SPaT information associated with the traffic light for a second predefined number of "recent" traffic signal cycles. The first predefined number may be substantially greater than the second predefined number.The system may include a trained machine module that may be trained using the historical traffic signal information and a supervised machine learning algorithm (e.g., supervised machine learning algorithm with Gaussian Process Regression (GPR)). The system may input the real-time traffic signal information to the trained machine module to output the predicted SPaT information for the future traffic signal cycle.In an example aspect, the SPaT information for the future traffic signal cycle may include an expected timing or duration for the green phase and the red phase for the traffic lights and an expected total duration for the future traffic signal cycle. In response to predicting the SPaT information for the future traffic signal cycle, the system may transmit the SPaT information to a vehicle.The vehicle may obtain the SPaT information from the system and may control vehicle movement based on the SPaT information and real-time vehicle information (e.g., a current vehicle speed toward the traffic light, a geolocation of the vehicle, etc.). For example, the vehicle may determine whether the vehicle may arrive at the traffic light in a green phase or red phase based on the SPaT information, the real-time vehicle information, and a geolocation of the traffic light. In response to determining that the vehicle at the current vehicle speed may not arrive at the traffic light in the green phase, the vehicle may modify (e.g., increase or decrease) the vehicle speed or maintain its current speed such that the vehicle may cross the traffic light in the green phase.The present disclosure discloses a system that assists a vehicle in crossing a traffic light in the green phase, thereby saving vehicle energy that may have been consumed if the vehicle had been waiting in the red phase at the traffic light. Further, the system predicts the SPaT information for the future traffic signal cycle based on both historical and real-time traffic signal information; therefore, the prediction is more accurate than the predictions output by conventional SPaT prediction systems.These and other advantages of the present disclosure are provided in detail herein.BRIEF DESCRIPTION OF THE DRAWINGSThe detailed description will be set forth with reference to the accompanying drawings. The use of the same reference numerals may indicate similar or identical elements. For various embodiments, elements and / or components other than those illustrated in the drawings may be used, and some elements and / or components may not be present in various embodiments. The elements and / or components in the figures are not necessarily drawn to scale. Throughout the disclosure, terms in the singular and plural may be used interchangeably depending on the context. FIG. 1 illustrates an example environment in which techniques and structures for providing the systems and methods disclosed herein may be implemented. FIG. 2 depicts a block diagram of an example traffic signal phase and timing (SPaT) prediction system and vehicle according to the present disclosure. FIG. 3 depicts a schematic diagram of a trained machine module according to the present disclosure. FIG. 4 depicts an example message displayed on a vehicle infotainment system, in accordance with the present disclosure. FIG. 5 depicts a flow diagram of an example method for predicting SPaT, in accordance with the present disclosure.DETAILED DESCRIPTIONThe disclosure will be described in more detail hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the disclosure are shown, and is not intended to be limiting.FIG. 1 illustrates an example environment 100 in which techniques and structures for providing the systems and methods disclosed herein may be implemented. The environment 100 may include a vehicle 102 that may travel on a road including one or more traffic lights 104 (or a traffic light 104). The vehicle 102 may take the form of any passenger or commercial vehicle, such as a car, a work vehicle, a crossover vehicle, a truck, a vans, a minivan, a taxi, a bus, etc. The vehicle 102 may be a manually driven vehicle and / or may be configured to operate in a semi- or fully autonomous mode, and may include any powertrain, such as a gasoline engine, one or more electrically actuated engine(s), a hybrid system, etc.The traffic light 104 may be configured to operate according to a fixed schedule, or may be dynamically actuated by a traffic light control server (not shown) based on traffic conditions, weather conditions, etc., on the road proximate the traffic light 104. In some aspects, when the traffic light 104 is operated according to a fixed schedule, actuation of a traffic signal phase (e.g., a green phase when the traffic light 104 emits green light, a yellow phase when the traffic light 104 emits yellow light, or a red phase when the traffic light 104 emits red light) may follow a fixed schedule. For example, if the traffic light 104 is operated according to a fixed schedule, the traffic light 104 may be in the green phase for 90 seconds, followed by 5 seconds in the yellow phase and then 60 seconds in the red phase. In this case, the timing or duration of each phase (e.g., 90 seconds for the green phase, 5 seconds for the yellow phase, and 60 seconds for the red phase) remains constant or fixed regardless of a flow of traffic or vehicles on the road. Furthermore, in this case, a total time duration of a "traffic signal cycle" can also remain constant. In some aspects, as described in the present disclosure, the traffic signal cycle means a cycle of traffic light operation beginning with the beginning of a green phase and ending at the beginning of a next green phase. Continuing with the example described below, the total amount of time for a traffic signal cycle associated with the traffic light 104 may be 155 seconds (i.e., a sum of 90+60+5 seconds) if the traffic light 104 may be operated according to a fixed schedule.On the other hand, if the traffic light 104 can be dynamically operated (e.g., by the traffic light control server) based on the traffic on the road, the operation of a traffic signal phase (i.e., the green phase, the red phase, or the yellow phase) may not follow a fixed pattern. Further, a time duration during which the traffic light 104 may be operated in each phase may not be fixed either. Moreover, in this case, the total time period of a traffic signal cycle may or may not be fixed. For the description of the present disclosure, it is assumed that the traffic light 104 is a dynamically operated traffic light. Further, for convenience, the red phase and the yellow phase are considered as the red phase.In addition to the vehicle 102 and the traffic light 104, the environment 100 may further include a traffic signal phase and timing prediction (SPaT) system 106 (system 106). The system 106 may be part of the vehicle 102 or may be a separate system hosted on a server (not shown). If the system 106 is hosted on a server, the system 106 may be communicatively coupled to the vehicle 102, as shown in FIG. 1.The system 106 may be an artificial intelligence / machine learning (AI / ML) based system that may be configured to predict traffic signal information associated with future traffic signal cycles for a plurality of traffic lights (including the traffic lights 104). Specifically, the system 106 may be an AI / ML-based system that may be configured to predict information associated with the SPaT for a future traffic signal cycle for the traffic light 104 based on historical traffic signal information and real-time traffic signal information associated with the traffic light 104. For example, the system 106 may predict a total amount of time, the actuation pattern of the phases, and the timing or time duration associated with each phase for the future (e.g., next) traffic signal cycle of the traffic light 104 based on historical and real-time traffic signal information associated with the traffic light 104. In response to predicting the SPaT information for the future traffic signal cycle, the system 106 may transmit the SPaT information to the vehicle 102. The vehicle 102 may use the SPaT information to increase or reduce vehicle speed, change heading, etc., such that the vehicle 102 may cross the traffic light 104 when the traffic light 104 is operated in the green phase where possible. In this way, the vehicle 102 may not be required to stop at the traffic light 104, thereby saving energy that may have been consumed by waiting at the traffic light 104.In some aspects, the system 106 may use the historical traffic signal information associated with the traffic light 104 as training data and a supervised machine learning algorithm (e.g., supervised machine learning algorithm with Gaussian Process Regression (GPR)) to generate a trained machine learning module. The historical traffic signal information may be SPaT information associated with the traffic light 104 for a predefined elapsed time period (e.g., the elapsed 3 months, 6 months, 12 months, etc.) or a first predefined number of elapsed traffic signal cycles (e.g., 10,000, 15,000, 25,000, etc.). As an example, the historical traffic signal information may include information associated with the total duration of each historical traffic signal cycle, timings or duration for each phase (green and red) in the historical traffic signal cycle, and / or the like. In response to generating the trained machine learning module, the system 106 may store the trained machine learning module in system memory (shown as system memory 222 in FIG. 2 ).The system 106 may be further configured to obtain the real-time traffic signal information from the traffic light control server and / or a traffic light controller (not shown) installed in the traffic light 104. In some aspects, the real-time traffic signal information may be SPaT information associated with the traffic light 104 for a second predefined number of "recent" traffic signal cycles (e.g., recent 5 or 10 traffic light cycles). As can be seen from the above description, the first predefined number of past traffic signal cycles may be substantially greater than the second predefined number of "recent" traffic signal cycles.In response to obtaining the real-time traffic signal information, the system 106 may execute instructions stored in the trained machine learning module and the real-time traffic signal information to predict the SPaT associated information for the future traffic signal cycle for the traffic lights 104. The system 106 may then transmit this information to the vehicle 102, which may control vehicle motion based on the received information. For example, if the vehicle 102 is an autonomous vehicle, the vehicle 102 may automatically control vehicle motion based on the received SPaT information such that the vehicle 102 may cross the traffic light 104 when the phase is possibly green. On the other hand, if the vehicle 102 is a manually driven vehicle or a semi-autonomous vehicle, the vehicle 102 may display a vehicle motion recommendation on a vehicle infotainment system (shown as infotainment system 216 in FIG. 2 ) based on the received SPaT information. The vehicle motion recommendation may include a recommendation to either increase or decrease the vehicle speed to assist the vehicle 102 in crossing the traffic lights 104 in the green phase. A vehicle operator (not shown) may view or hear the vehicle motion recommendation on the vehicle infotainment system and may control vehicle motion accordingly.Further details regarding the system 106 and the vehicle 102 are described below in connection with FIG. 2.The vehicle 102, the system 106, and the vehicle operator implement and / or perform operations as described herein in the present disclosure according to user manual and safety policies. Additionally, for each action taken by the vehicle or vehicle operator, all rules specific to the location and operation of the vehicle 102 (e.g., federal, state, country, city, etc.) should be observed. The notifications or recommendations as provided by the vehicle 102 or the system 106 should be treated as suggestions and followed only according to any rules specific to the location and operation of the vehicle 102.FIG. 2 depicts a block diagram of an example traffic signal phase and timing (SPaT) prediction system 200 (system 200) and a vehicle 202, in accordance with the present disclosure. The system 200 may be the same as the system 106 and the vehicle 202 may be the same as the vehicle 102 described above in connection with FIG. 1. During the description of FIG. 2, reference is made to FIGS. 3 and 4.The system 200 and the vehicle 202 may be communicatively coupled to each other via one or more networks 204 (or a network 204). The system 200 and the vehicle 202 may further communicatively couple to one or more servers 206 (or a server 206) via the network 204. In some aspects, the server 206 may be part of a cloud-based computing infrastructure and associated with and / or include a telematics service delivery network (SDN) that provides digital data services to the vehicle 202 and other vehicles (not shown in FIG. 2 ) that may be part of a commercial fleet of vehicles. In further aspects, the server 206 may be a traffic light control server that may be configured to store historical traffic signal information associated with a plurality of traffic lights (including the traffic light 104) and transmit the historical traffic signal information to the system 200 at a predefined frequency or when the system 200 transmits a request to the server 206 to retrieve / acquire the historical traffic signal information. The server 206 may be further configured to determine real-time traffic signal information associated with the plurality of traffic lights (including the traffic lights 104) and transmit the real-time traffic signal information to the system 200 when the system 200 transmits a request to the server 206 to retrieve / acquire the real-time traffic signal information. The concept of historical traffic signal information and real-time traffic signal information has already been described above in connection with FIG. 1. In some aspects, the server 206 may transmit the historical traffic signal information and the real-time traffic signal information associated with the traffic light 104 to the system 200 via the network 204. Network 204 illustrates an example communication infrastructure in which the connected devices discussed in various embodiments of this disclosure may communicate. The network 204 may be and / or include the Internet, a private network, a public network, or other configuration that operates using any one or more of any known communication protocols, such as transmission control protocol / Internet protocol (TCP / IP), Bluetooth ®, BLE, Wi-Fi based on the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard, ultra-wideband (UWB), and cellular technologies such as time division multiple access (TDMA), code division multiple access (CDMA), high-speed packet access (HSPDA), for example, Long-term evolution (LTE), Global System for Mobile communications (GSM), and Fifth Generation (5G), to name a few examples.The vehicle 202 may include a plurality of entities including, but not limited to, a vehicle transceiver 208, a vehicle memory 210, a vehicle processor 212, a vehicle control unit (VCU) 214, an infotainment system 216, and / or the like, which may be communicatively coupled to one another. The vehicle transceiver 208 may be configured to receive / transmit data / information / signals from external systems and devices over the network 204. For example, the vehicle transceiver 208 may transmit information / signals to or receive information / signals from the system 200 and / or the server 206. The vehicle transceiver 208 may be further configured to transmit / receive signals / information to and from the vehicle units described above.The VCU 214 may include a plurality of electronic control units (ECUs, not shown) and / or a plurality of vehicle sensors (not shown), which may enable the VCU 214 to control operating aspects of the vehicle and implement one or more instruction sets received from the vehicle processor 212. For example, the VCU 214 may control vehicle speed, direction of motion of the vehicle, etc., if the vehicle 202 may be a fully or semi-autonomous vehicle. The VCU 214 may be further configured to determine real-time vehicle information. In some aspects, the real-time vehicle information may include a real-time (or current) vehicle speed, a vehicle direction of motion, a geolocation of the vehicle, and / or the like.The vehicle processor 212 may be arranged in communication with one or more storage devices arranged in communication with the respective computing systems (e.g., the vehicle memory 210 and / or one or more external databases not shown in FIG. 2 ). The vehicle processor 212 may use the vehicle memory 210 to store programs in code and / or store data for performing aspects according to the disclosure. The vehicle memory 210 may be a non-transitory computer readable storage medium or memory storing program code / s that may enable the vehicle processor 212 to perform operations according to the present disclosure. The vehicle memory 210 may include any one or combination of volatile memory elements (e.g., dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), etc.), and may include any one or more nonvolatile memory elements (e.g., erasable programmable read-only memory (EPROM), flash memory, electronically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), etc.).The infotainment system 216 may include a touch screen interface portion and may include voice recognition features, biometric identification capabilities that may identify users based on face recognition, voice recognition, fingerprint identification, or other biological identifiers. In other aspects, the infotainment system 216 may be further configured to receive user instructions via the touch screen interface portion and / or output or display notifications (including visual and / or audible warning notifications), navigation maps, recommendations, messages, etc., on the touch screen interface portion.The system 200 may include a plurality of units including, but not limited to, a system transceiver 218, a system processor 220, and a system memory 222. The system transceiver 218 may be configured to receive / transmit information / signals from or to external systems or devices via the network 204. For example, the system transceiver 218 may receive / transmit information / signals from the vehicle 202 and the server 206.The system processor 220 may utilize the system memory 222 to store programs in code and / or store data for performing operations in accordance with the present disclosure. System memory 222 may be a non-transitory computer readable storage medium or memory storing program code / s that may enable system processor 220 to perform the operation according to the present disclosure. In some aspects, system memory 222 may include a plurality of databases and modules including, but not limited to, a training data database 224 (or training data 224), a real-time traffic signal information database 226 (or real-time traffic signal database 226), a machine learning module 228, and a trained machine module 230. The functions of these storage databases and modules may be understood in connection with the description provided below.In operation, the system transceiver 218 may receive the historical traffic signal information associated with the traffic light 104 from the server 206 via the network 204. In some aspects, the system transceiver 218 may receive the historical traffic signal information from the server 206 when the system processor 220 transmits, via the system transceiver 218, a request to retrieve / obtain the historical traffic signal information to the server 206. The system transceiver 218 may additionally receive a geolocation of the traffic lights 104 from the server 206. The system transceiver 218 may transmit the geolocation of the traffic lights to the system memory 222 for storage purposes.In response to obtaining the historical traffic signal information associated with the traffic light 104, the system transceiver 218 may transmit the historical traffic signal information to the training data 224 for storage purposes. In other words, the training data 224 may include the historical traffic signal information associated with the traffic light 104.In some aspects, the system processor 220 may be an AI / ML-based processor that may generate the trained machine module 230 by using the training data 224 (i.e., the historical traffic signal information) and a supervised machine learning algorithm stored in advance in the machine learning module 228. In an example aspect, the supervised machine learning algorithm may be a supervised Gaussian regression machine learning (GPR) algorithm. In response to generating the trained machine module 230, the system processor 220 may execute the instructions stored in the trained machine module 230 to "predict" traffic signal information or SPaT information for a future traffic signal cycle associated with the traffic light 104 by using the real-time traffic signal information as input. In other words, if the trained machine module 230 can be generated, the system processor 220 can obtain the real-time traffic signal information from the server 206 and input the real-time traffic signal information to the trained machine module 230 to predict the SPaT information for a future traffic signal cycle.Those of ordinary skill in the art can appreciate that machine learning is an application of artificial intelligence (AI), using which systems or processors (e.g., system processor 220) may have the ability to automatically learn from experiences and improve without being explicitly programmed. Machine learning focuses on using data and algorithms to mimic how people learn. In some aspects, machine learning algorithms may be generated to generate classifications and / or predictions. Machine learning based systems may be used for a variety of applications including, but not limited to, speech recognition, image or video processing, and statistical analysis, probability determination, traffic signal phase and timing determination, and / or the like.Machine learning may be of various types based on data or signals available to the learning system. For example, the machine learning approach may include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. Supervised learning is an approach that can be supervised by a human. In this approach, the machine learning algorithm may use labeled training data (e.g., training data 224) and defined variables. In the case of supervised learning, both the input and the output of the algorithm may be predetermined / defined, and the algorithms may be trained to classify data and / or accurately predict results.Generally, supervised learning may be one of two types: "regression" and "classification.". In classification learning, the learning algorithm may assist in dividing the dataset into classes based on different parameters. In this case, a computer program may be trained on the training dataset and the computer program may categorize input data into different classes based on the training. Some known methods used in classification learning include logistic regression, K-nearest neighbor, support vector machines (SVM), kernel SVM, naive bayes, decision tree classification, and random forest classification.In regression learning, the learning algorithm may predict an output value that may be continuous in nature or a real value. Some known methods used in regression learning include simple linear regression, multiple linear regression, polynomial regression, support vector regression, decision tree regression, and random forest regression. In the present disclosure, the machine learning module 228 may be based on a Gaussian Regression Machine Learning (GPR) algorithm, and the system processor 220 may execute the GPR Supervised Machine Learning algorithm stored in the machine learning module 228 and the training data 224 (i.e., the historical traffic signal information) to generate the trained machine module 230.Unsupervised learning is an approach involving algorithms that can be trained with unlabeled data. An unsupervised learning algorithm may analyze the data itself and find patterns in input data. Further, semi-supervised learning is a combination of supervised learning and unsupervised learning. A semi-supervised learning algorithm includes labeled training data; however, the semi-supervised learning algorithm may still find patterns in the input data. Reinforcement learning is a multistage or dynamic process. This model is similar to supervised learning, but cannot be trained using example data. This model can learn "on-the-fly" by using trial and error. In reinforcement learning, a sequence of successful results may be reinforced to develop the best recommendation or policy for a given problem.As described above, in the present disclosure, the system processor 220 uses a regression-based algorithm for supervised learning to generate the trained machine module 230 by using the training data 224. In other words, the trained machine module 230 may be trained using the training data 224. The system processor 220 may continuously modify and update the trained machine module 230 as new training data may be received by the system 200 to make the trained machine module 230 more accurate over time. In response to generating the trained machine module 230, the system processor 220 may store the trained machine module 230 in the system memory 222.An example schematic of the trained machine module 230 is depicted in FIG. 3. SPaT information is depicted as a traffic light symbol in FIG. 3. In an example aspect, as shown in FIG. 3, the trained machine module 230 may be trained by using multiple sets 302 a, 302 b, 302 nof training data or historical traffic signal information. Specifically, during the "training" or generating of the trained machine module 230, the system processor 220 may execute the machine learning module 228 to "predict" SPaT information 304 associated with a future traffic signal cycle by using the set 302 aand then compare the predicted SPaT information 304 with actual SPaT information. Based on the comparison, the system processor 220 may determine an accuracy level of the machine module and fine tune the module to improve accuracy. Further, in response to fine tuning, the system processor 220 may likewise cause the machine learning module 228 to predict SPaT information 306, 308 for future traffic signal cycles by using the sets 302 b, 302 nand perform the comparison as described above until the accuracy level of the machine module may be above a predefined threshold. In response to determining that the level of accuracy is above the predefined threshold, the system processor 220 may treat the machine module as "trained" and store the trained machine module 230 in the system memory 222, as described above.In response to generating and storing the trained machine module 230, the system processor 220 may obtain, via the system transceiver 218, the real-time traffic signal information associated with the traffic light 104 from the server 206 and / or a traffic light controller installed in the traffic light 104 via the network 204. Specifically, in this case, the system processor 220 may transmit, via the system transceiver 218, a request to the server 206 and / or the traffic light controller to obtain the real-time traffic signal information. In response to transmitting the request, the system transceiver 218 may receive the real-time traffic signal information and transmit the received information to the system processor 220 and the real-time traffic signal database 226 for storage purposes.In response to obtaining the real-time traffic signal information from the system transceiver 218, the system processor 220 may execute instructions stored in the trained machine module 230 to predict traffic signal information (i.e., SPaT information) associated with a future traffic signal cycle for the traffic lights 104 based on the real-time traffic signal information. For example, as shown in FIG. 3, the system processor 220 may input real-time traffic signal information 310 (e.g., SPaT information for past 8 or 10 traffic signal cycles) associated with the traffic light 104 to the trained machine module 230 and execute the instructions stored in the trained machine module 230 to predict SPaT information 312 associated with a future (or next) traffic signal cycle. As an example, the system processor 220 may input the timing and duration associated with the green phase of the traffic lights 104 for the past 10 traffic signal cycles to the trained machine module 230 and output the predicted or estimated timing and duration associated with the green phase of the traffic lights 104 for a future (e.g., next) traffic signal cycle.An example mathematical expression associated with the trained machine module 230 is illustrated below. The mathematical expression should not be construed as limiting the scope of the present disclosure and is provided as an exemplary expression that may be associated with the trained machine module 230. Any other mathematical expression may also be associated with the trained machine module 230 without departing from the scope of the present disclosure.A Gaussian process (associated with trained machine module 230) may be defined as: f(x)~GP[m(x ), k(x, x')], where m(x) may be an average function and k(x, x') may be a covariance function for a pair (x, x'). Further, in an exemplary aspect, m(x)=E[f(x)] and k(x, x')=E[(f(x)-m(x))(f(x')-m(x'))] apply, where E[X] denotes an expected value of X.The example steps performed by the system processor 220 and / or the trained machine module 230 to model the SPaT information with Gaussian process regression may include a first step in which the model or trained machine module 230 may be formed (or "model formulation") in which y=f(x), f(x)~GP[m(x), k(x, x'|θ)], where y is the model and θ is the model parameter vector. The model parameter θ can be estimated with maximum likelihood.As a second step, the nth duration d n( e.g., the duration of the green phase), may be predicted according to the duration of the last "k" durations (d n-k ~ d n-1). In some aspects, the input to the model x may be n= [ d n-k.... d n-k+1... d n-1] and the output from the model n= D n. Further, as described above, the model may be trained with collected historical SPaT data D (or training dataset D), where D=[X, Y]. Here, X - [x 1, x 2, x 3....] and Y = [y 1, y 2, y 3....].In response to predicting the SPaT information 312 associated with the future traffic signal cycle, the system processor 220 may output / transmit the SPaT information 312 to the vehicle transceiver 208 via the network 204. The system processor 220 may further transmit, via the system transceiver 218, the geolocation of the traffic lights 104 (which the system 200 may have received from the server 206 as described above) to the vehicle transceiver 208.The vehicle transceiver 208 may receive the SPaT information 312 and the geolocation of the traffic lights from the system transceiver 218, and may transmit the received information to the vehicle memory 210 and to the vehicle processor 212 for storage purposes. The vehicle processor 212 may obtain the SPaT information 312 and the geolocation of the traffic lights from the vehicle transceiver 208 or the vehicle storage 210, and may also obtain the real-time vehicle information from the VCU 214. As described above, the real-time vehicle information may include a real-time (or current) vehicle speed, a vehicle direction of motion, a geolocation of the vehicle, and / or the like. In response to obtaining the information described above, the vehicle processor 212 may correlate the SPaT information 312 and the geolocation of the traffic lights with the real-time vehicle information to determine whether the vehicle 202 will cross the traffic lights 104 in the green phase if the vehicle 202 continues to travel at the current speed (or in the same direction of travel). In response to determining that the vehicle 202 may not cross the traffic light 104 in the green phase based on the correlation, the vehicle processor 212 may generate a command signal. The vehicle processor 212 may then output / transmit the generated command signal to the infotainment system 216 and / or the VCU 214. In response to obtaining the command signal from the vehicle processor 212, the infotainment system 216 may output a predefined message based on the command signal (audible or visual). For example, as shown in view 402 of FIG. 4, infotainment system 216 may display "reduce vehicle speed to pass the next traffic light in the green phase.". In response to the vehicle operator (not shown) listening or seeing the message, he may reduce the vehicle speed (while considering the traffic present and considering the local rules and laws) to cross the traffic lights 104 in the green phase. In this manner, the system 200 assists the vehicle 202 in crossing the traffic light 104 in the green phase, thereby saving vehicle energy that may have been consumed if the vehicle 202 had been waiting at the traffic light 104 in the red phase. Although the above example describes an aspect in which the command signal indicates to the vehicle operator to reduce the vehicle speed, in some aspects, the command signal may indicate to the vehicle operator to increase the vehicle speed and / or change the vehicle direction of motion (e.g., take a turn-around) to assist the vehicle 202 in crossing the traffic lights 104 in the green phase while considering the traffic present and considering the local rules and laws. In further aspects, if the vehicle 202 may be a fully or semi-autonomous vehicle, the VCU 214 may autonomously control vehicle motion (e.g., decrease or increase vehicle speed or change vehicle direction of motion while considering all local rules and laws) in response to obtaining the command signal from the vehicle processor 212 such that the vehicle 202 may cross the traffic lights 104 in the green phase. In some embodiments, the vehicle may know certain local traffic laws, such as speed limit, and take these into account in generating the output. For example, the system 200 may incorporate a maximum allowable speed into the output or recommend a decrease in speed if the vehicle will not be able to pass the next traffic light in a green phase without exceeding the local speed limit.Although the above description describes an aspect in which the system processor 220 generates the trained machine module 230 and predicts the SPaT information 312 for the traffic lights 104 by executing the instructions stored in the trained machine module 230, it will be appreciated by those of ordinary skill in the art that the system processor 220 may generate similar trained machine modules for a plurality of traffic lights (one trained machine module for each traffic light). The system processor 220 may then determine SPaT information associated with each traffic light on a travel route of the vehicle 202, and transmit the determined SPaT information to the vehicle 202. The vehicle 202 may then control vehicle motion as described above based on the SPaT information such that the vehicle 202 may cross most (or all) of the traffic lights on the vehicle's route in respective green phases.FIG. 5 depicts a flow diagram of an example method 500 for predicting SPaT, in accordance with the present disclosure. FIG. 5 may be described with continued reference to previous figures. The following process is exemplary and is not limited to the steps described below. Moreover, alternative embodiments may include more or fewer steps than shown or described herein, and these steps may include in an order that differs from the order described in the following embodiments.The method 500 begins at step 502. At step 504, the method 500 may include obtaining, by the system processor 220, the real-time traffic signal information from the server 206 or the traffic light controller. At step 506, the method 500 may include executing instructions stored in the trained machine module 230 by the system processor 220 to predict the SPaT 312 based on the real-time traffic signal information. At step 508, the method 500 may include outputting, by the system processor 220, the SPaT 312 to the vehicle 202. The method 500 may end at step 510.In the foregoing disclosure, reference has been made to the accompanying drawings, which form a part hereof, and illustrate specific implementations in which the present disclosure may be practiced. It should be understood that other implementations may be utilized and structural changes may be made without departing from the scope of the present disclosure. References in the specification to "an embodiment," "an embodiment," etc., indicate that the described embodiment may include one(s) specific feature, structure, or characteristic, but each embodiment does not necessarily have to include that(s) specific feature, structure, or characteristic. Furthermore, such formulations do not necessarily relate to the same embodiment. Further, when a feature(s) structure, or characteristic is described in connection with one embodiment, those skilled in the art will recognize such feature(s) structure, or characteristic in connection with other embodiments whether or not explicitly described. Further, the functions described herein may be performed in one or more of hardware, software, firmware, digital components, or analog components, as appropriate. For example, one or more application specific integrated circuits (ASICs) may be programmed to execute one or more of the systems and procedures described herein. Certain terms used throughout the specification and claims refer to specific system components. It will be appreciated by those skilled in the art that the components may be designated by other terms. In this document, it is not intended to distinguish between components which differ according to the designation, but not in terms of their function.Also, it should be understood that the word "example" as used herein is intended to be non-exclusive and non-limiting. In particular, the word "example" as used herein indicates one of several examples, and it is understood that no undue emphasis or preference is directed to the particular example described.A computer readable medium (also referred to as a processor readable medium) includes any non-transitory (e.g., tangible) medium that participates in providing data (e.g., instructions) that may be read by a computer (e.g., by a processor of a computer). Such a medium may take many forms, including, but not limited to, non-volatile media and volatile media. Computing devices may include computer-executable instructions, where the instructions may be executable by one or more computing devices, such as those listed above, and may be stored on a computer-readable medium.With respect to the processes, systems, methods, heuristics, etc. described herein, it should be understood that, although the steps of such processes, etc. have been described as occurring according to a certain ordered sequence, in practice, such processes could be practiced with the described steps performed in an order that varies from the order described herein. Further, it should be understood that certain steps could be performed simultaneously, that other steps could be added, or that certain steps described herein could be omitted. In other words, the descriptions of processes herein are for the purpose of illustrating various embodiments and should not be construed as limiting the claims. Accordingly, it is to be understood that the foregoing description is intended to be illustrative and not restrictive. From reading the foregoing description, many embodiments and applications other than the examples provided will be apparent. The scope should be determined, not with reference to the foregoing description, but should instead be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. It is to be understood and intended that there will be future developments in the art discussed herein, and that the disclosed systems and methods will be incorporated into such future embodiments. Overall, it is to be understood that the application may be modified and varied.All terms used in the claims are intended to have their general meaning as known to those skilled in the art of technologies described herein, unless expressly stated to the contrary in this specification. In particular, the use of the singular articles such as "a", "an", "the", "the", etc. is to be understood to mean one or more of the stated elements unless a claim gives an explicit limitation to the contrary. With phrases expressing conditional contexts, such as, but not limited to, "may," "could," "may," or "may in some circumstances," it is generally intended to convey that certain embodiments may include certain features, elements, and / or steps, while other embodiments may not include these unless specifically stated otherwise or otherwise will be apparent from the context used. Thus, such phrases expressing conditional relationships are generally not intended to imply that features, elements, and / or steps are required in any way for one or more embodiments.In one aspect of the invention, each of the historical traffic signal information, the real-time traffic signal information, and the traffic signal information associated with the future traffic signal cycle includes information associated with a traffic signal phase and timing (SPaT) associated with each traffic signal cycle.In one aspect of the invention, the historical traffic signal information is associated with a second predefined number of traffic signal cycles, and wherein the second predefined number of traffic signal cycles is greater than the first predefined number of recent traffic signal cycles.In one aspect of the invention, the method includes generating the trained machine model by using the training data and a supervised machine learning algorithm with Gaussian Process Regression (GPR).In one aspect of the invention, acquiring the real-time traffic signal information includes acquiring the real-time traffic signal information from a traffic light control server.In one aspect of the invention, outputting the traffic signal information includes outputting the traffic signal information to a vehicle.

Claims

A system, comprising: a system transceiver configured to: receive historical traffic signal information associated with a traffic light; and receive real-time traffic signal information associated with the traffic light for a first predefined number of recent traffic signal cycles; a system memory configured to store training data and a trained machine model, wherein the trained machine model is trained using the training data comprising the historical traffic signal information; a system processor communicatively coupled to the system transceiver and the system memory, wherein the system processor is configured to: obtain the real-time traffic signal information from the system transceiver; executing instructions stored in the trained machine model to predict traffic signal information associated with a future traffic signal cycle based on the real-time traffic signal information; and output the traffic signal information associated with the future traffic signal cycle.The system of claim 1, wherein each of the historical traffic signal information, the real-time traffic signal information, and the traffic signal information associated with the future traffic signal cycle comprises information associated with a traffic signal phase and timing (SPaT) associated with each traffic signal cycle.The system of claim 1, wherein the historical traffic signal information is associated with a second predefined number of traffic signal cycles, and wherein the second predefined number of traffic signal cycles is greater than the first predefined number of recent traffic signal cycles.The system of claim 1, wherein the system processor is further configured to generate the trained machine model by using the training data and a supervised Gaussian regression machine learning (GPR) algorithm.The system of claim 1, wherein the system transceiver receives the historical traffic signal information and the real-time traffic signal information from a traffic light control server.The system of claim 1, wherein the system processor outputs the traffic signal information associated with the future traffic signal cycle to a vehicle.A vehicle, comprising: a vehicle transceiver configured to receive traffic signal information associated with a future traffic signal cycle for a traffic light from a server, wherein the traffic signal information associated with the future traffic signal cycle is based on historical traffic signal information associated with the traffic light and real-time traffic signal information associated with the traffic light for a first predefined number of recent traffic signal cycles; a vehicle controller configured to determine real-time vehicle information; and a vehicle processor communicatively coupled to the vehicle transceiver and the vehicle control unit, the vehicle processor configured to: obtain the traffic signal information associated with the future traffic signal cycle from the vehicle transceiver and the real-time vehicle information from the vehicle control unit; generate a command signal based on the traffic signal information and the real-time vehicle information; and output the command signal.The vehicle of claim 7, wherein the historical traffic signal information is associated with a second predefined number of traffic signal cycles, wherein the second predefined number of traffic signal cycles is greater than the first predefined number of recent traffic signal cycles.The vehicle of claim 7, wherein each of the historical traffic signal information, the real-time traffic signal information, and the traffic signal information associated with the future traffic signal cycle comprises information associated with a traffic signal phase and timing (SPaT) associated with each traffic signal cycle.The vehicle of claim 7, wherein the traffic signal information associated with the future traffic signal cycle is determined using a Gaussian Process Regression (GPR) supervised machine learning algorithm.The vehicle of claim 7, wherein the real-time vehicle information comprises a current vehicle speed and a direction of vehicle motion.The vehicle of claim 7, wherein the vehicle processor is further configured to: correlate the traffic signal information associated with the future traffic signal cycle with the real-time vehicle information; and generate the command signal based on the correlation.The vehicle of claim 7, wherein the vehicle processor outputs the command signal to a vehicle infotainment system, and wherein the vehicle infotainment system outputs a predefined message in response to receiving the command signal.The vehicle of claim 7, wherein the vehicle processor outputs the command signal to the vehicle control unit, wherein the vehicle control unit autonomously controls a vehicle motion in response to receiving the command signal.A method, comprising: obtaining, by a processor, real-time traffic signal information, wherein the real-time traffic signal information is associated with a traffic light for a first predefined number of recent traffic signal cycles; executing, by the processor, instructions stored in a trained machine model to predict traffic signal information associated with a future traffic signal cycle based on the real-time traffic signal information, wherein the trained machine model is trained using training data comprising historical traffic signal information associated with the traffic light; and outputting, by the processor, the traffic signal information associated with the future traffic signal cycle.

Images