Hybrid distributed prediction of traffic signal state changes

• US 10,008,113 B2
• Filed: 06/10/2016
• Issued: 06/26/2018
• Est. Priority Date: 04/12/2013
• Status: Active Grant

• Current Assignee: Traffic Technology Services, Inc.
• Original Assignee: Traffic Technology Services, Inc.
• Inventors: Ova, Kiel Roger, Bauer, Thomas, Ma, Jingtao, Hatcher, Kyle Zachary
• Primary Examiner(s): Bendidi, Rachid
• Application Number:
  US15/179,850
• Publication Number:
  US 20160284215A1
• Time in Patent Office: 746 Days
• CPC Class Codes: G06V 20/584 of vehicle lights or traffic lights
• CPC Class Codes: G08G 1/0116 from roadside infrastructure, e.g. beacons
• CPC Class Codes: G08G 1/0129 for creating historical data or processing based on historical data
• CPC Class Codes: G08G 1/0141 for traffic information dissemination
• CPC Class Codes: G08G 1/09623 Systems involving the acquisition of information from passive traffic signs by means mounted on the vehicle G08G1/0967 takes precedence
• CPC Class Codes: G08G 1/096716 where the received information does not generate an automatic action on the vehicle control
• CPC Class Codes: G08G 1/096741 where the source of the transmitted information selects which information to transmit to each vehicle
• CPC Class Codes: G08G 1/096758 where no selection takes place on the transmitted or the received information
• CPC Class Codes: G08G 1/096775 where the origin of the information is a central station

Abstract

Computer-implemented predictions of upcoming traffic control signal states or state changes can be used to improve driver convenience, safety, and fuel economy. Such information can be used advantageously by a human operator, or by an autonomous or semi-autonomous vehicle control system. Predictions can be computed with suitable machines installed in a vehicle, in cooperation with a remote back-end server system. The prediction computations in the vehicle may be supported by data communicated to the vehicle computing machinery over various wireless communications, including telecom systems, DSRC, etc.

Claims

1. An apparatus for use in a vehicle for traffic signal predictions, comprising:

a processor coupled to a memory to execute non-transient, machine-readable instructions stored in the memory;

a first interface coupled to the processor for wireless data communications over a network with a remote back-end server;

a receiver coupled to the processor to receive a current state of a target traffic signal;

the stored instructions configured to;

implement a signal state prediction process;

when the vehicle is approaching or near a traffic signal, send a request message via the first interface to the remote back-end server, the request message including information to identify the traffic signal as a target traffic signal;

receive a reply message from the remote back-end server, the reply message including statistical data based on historical operation of the target traffic signal, wherein the back-end server generates the statistical data by acquiring call data and signal status data provided by a field traffic signal controller (FSC) associated with the target traffic signal during a selected collection period, and storing the acquired call data in a database coupled to the-remote-back-end server;

input the statistical data to the signal state prediction process;

commence executing the signal state prediction process wherein

executing the signal state prediction process includes;

identifying a signal status at a last sync point of the FSC, initializing the signal state prediction process to an initial state, advancing the signal state prediction process from the initial state to a second state, the second state corresponding to the last sync point of the FSC, and further advancing the signal state prediction process from the second state, based on the statistical data in the reply message, for a time period from the last sync point to a current time state so that at the current time the FSC and the signal state prediction process running in the vehicle are synchronized to the current time state;

receive the current state of the target traffic signal via the receiver;

in the signal state prediction process, update a state of the target traffic signal to match the current state of the target traffic signal;

in the signal state prediction process, generate a prediction of a next state change of the target traffic signal in the future, by;

fast forwarding the signal state prediction process from the current time state, based on the received statistical data, terminating the signal state prediction process at a future time state, and

generating a prediction of future signal status of the target traffic signal based on the future time state of the emulator;

and then transmit the generated prediction to an on-board network in the vehicle for display in the vehicle or input to engine control units or other on-board computers for use in controlling operation of the vehicle.

2. The apparatus of claim 1 including an electronic display in the vehicle, wherein the electronic display is coupled to the on-board network to display information based on the prediction.

3. The apparatus of claim 1 wherein the receiver is configured to receive a short-range broadcast message from a traffic signal controller, the broadcast message indicating a current state of a traffic signal operatively coupled to the traffic signal controller.

4. The apparatus of claim 3 wherein the short-range broadcast message comprises a Dedicated Short-Range Communications system (DSRC) compliant message.

5. The apparatus of claim 1 wherein the receiver is configured to receive image data from a camera in the vehicle and to determine the current state of a traffic signal in front of the vehicle based on the acquired image data.

6. The apparatus of claim 5 wherein the camera is built into the vehicle.

7. The apparatus of claim 1 wherein the receiver is configured to receive image data from a camera in the vehicle and to determine a current state of a traffic signal head in view of the camera based on the acquired image data.

8. A vehicle comprising:

a processor coupled to a memory to execute non-transient, machine-readable instructions stored in the memory, the stored instructions including instructions to implement a signal state prediction process;

a first interface coupled to the processor for wireless data communications over a network with a remote back-end server;

a camera coupled to the processor for capturing image data including an image of a target traffic signal located in front of the vehicle;

the stored instructions configured to;

when the vehicle is approaching or near a traffic signal, send a request message via the first interface to the remote back-end server, the request message including information to identify the traffic signal as a target traffic signal;

receive a reply message from the remote back-end server, the reply message including statistical data based on historical operation of the target traffic signal;

input the statistical data to the signal state prediction process;

begin executing the signal state prediction process, starting at a last sync point for the target traffic signal, and informed by the statistical data, advance the signal state prediction process to catch up to a current real-world time by clocking it at a clock rate faster than real world time, so that at the current time a field traffic signal controller (FSC) associated with the target traffic signal and the signal state prediction process running in the vehicle are synchronized to the same state;

process image data captured by the camera to determine a current state of the target traffic signal;

input the current state of the target traffic signal to the signal state prediction process;

in the signal state prediction process, update a state of the target traffic signal to match the current state of the target traffic signal;

in the signal state prediction process, generate a prediction of a next state change of the target traffic signal in the future, by;

fast forwarding the signal state prediction process from the current state, by clocking the signal state prediction process at a clock rate faster than real-world time, and then terminating the signal state prediction process at a selected future time state, and

generating a prediction of a future state of the target traffic signal based on the future time state of the signal state prediction process; and

transmit the prediction to an on-board network for display in the vehicle or input to engine control units or other on-board computers for use in controlling operation of the vehicle.

9. The vehicle of claim 8 wherein the stored instructions are further configured to:

generate a warning message based on the prediction; and

transmit the warning message for display in the vehicle.

10. The vehicle of claim 8 wherein the stored instructions are further configured to:

generate an audio warning message based on the prediction; and

transmit the audio warning message to be played over an audio output in the vehicle.

11. The vehicle of claim 8 wherein the stored instructions are further configured to store the prediction in a datastore.

12. The vehicle of claim 8 wherein the statistical data includes:

an observed maximum green signal duration, MAX_g_obs; and

for each second in a set [0,MAX_g_obs] in each phase, an empirical probability p for vehicle arrival/presence.

13. A method comprising:

loading and executing an emulator process in a processor in a vehicle to emulate operation of a field traffic signal controller (FSC) at a physical location;

at a back-end server system capable of wireless communication with the vehicle, acquiring call data and signal status data provided by the FSC during a selected collection period, and storing the acquired call data in a database;

at the back-end server system, predicting future detection data for the FSC based on a statistical analysis of the acquired call data stored in the database;

in the processor in the vehicle, as the vehicle is approaching the location of the FSC, sending a request message to the back-end server, the request message identifying the FSC;

in the processor in the vehicle, receiving a reply message from the back-end server, the reply message comprising prediction system input data for the FSC, the input data including the predicted future detection data to predict likely upcoming state changes of the FSC;

in the processor in the vehicle, identifying a signal status at a last sync point of the FSC in the database;

in the processor in the vehicle, initializing the emulator process to an initial state;

in the processor in the vehicle, advancing the emulator process from the initial state to a second state, the second state corresponding to a last sync point of the FSC;

in the processor in the vehicle, further advancing the emulator process from the second state, based on the prediction system input data in the received response, for a time period from the last sync point to a current time, so that at the current time the FSC and the emulator process running in the vehicle are synchronized to a current time state;

in the processor in the vehicle, receiving real-time state information from the FSC;

in the processor in the vehicle, adjusting a current state of the emulator process to match a current state of the FSC based on the real-time state information;

in the processor in the vehicle, fast forwarding the emulator process from the adjusted current state, based on the predicted future detection data, by clocking the emulator process at a rate faster than real-world time, so that it “

gets ahead”

of the current state of the FSC;

in the processor in the vehicle, terminating the emulator process at a future time state;

in the processor in the vehicle, predicting a signal status based on the future time state of the emulator process; and

in the processor in the vehicle, communicating a result based on the predicted signal status to a network on board the vehicle for display in the vehicle or input to engine control units or other on-board computers for use in controlling operation of the vehicle.

14. The method of claim 13 wherein the result includes an indication of the predicted signal status and a predicted time interval remaining until the FSC changes state to the predicted signal status.

15. The method of claim 13 wherein receiving real-time state information from the FSC comprises receiving a short-range broadcast message from the FSC.

16. The method of claim 15 wherein the short-range broadcast message comprises a Dedicated Short-Range Communications system (DSRC) compliant message.

17. The method of claim 13 wherein receiving real-time state information includes receiving image data from a camera in the vehicle determining the current state of the target traffic signal based on the image data.

18. The method of claim 17 wherein the camera is mounted to the vehicle.