WO2019074478A1 - Autonomous safety systems and methods for vehicles - Google Patents

Autonomous safety systems and methods for vehicles Download PDF

Info

Publication number
WO2019074478A1
WO2019074478A1 PCT/US2017/055751 US2017055751W WO2019074478A1 WO 2019074478 A1 WO2019074478 A1 WO 2019074478A1 US 2017055751 W US2017055751 W US 2017055751W WO 2019074478 A1 WO2019074478 A1 WO 2019074478A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
control system
output value
safety control
based
Prior art date
Application number
PCT/US2017/055751
Other languages
French (fr)
Inventor
Vivek Anand Sujan
Original Assignee
Vivek Anand Sujan
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Vivek Anand Sujan filed Critical Vivek Anand Sujan
Priority to PCT/US2017/055751 priority Critical patent/WO2019074478A1/en
Publication of WO2019074478A1 publication Critical patent/WO2019074478A1/en

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/0088Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2201/00Application
    • G05D2201/02Control of position of land vehicles
    • G05D2201/0213Road vehicle, e.g. car or truck

Abstract

A safety control system (10) is provided for a vehicle (12), and includes a plurality of sensors (24) associated with the vehicle (12) configured to generate data related to at least one environmental characteristic. A system model generation unit (30) receives the data related to the at least one environmental characteristic, and to generate a system model based on the data. The system model represents a current environmental status of the vehicle (12) based on the at least one environmental characteristic. A learning unit (32) receives the system model and learn the system model through a plurality of executions using the data related to the at least one environmental characteristic for performing a risk assessment of an environmental familiarity of the system model, and generates an output value that determines at least one operating parameter of the vehicle (12) such that the vehicle (12) is operated based on the operating parameter.

Description

AUTONOMOUS SAFETY SYSTEMS AND METHODS FOR VEHICLES

FIELD OF THE DISCLOSURE

[0001] The present disclosure generally relates to vehicle control systems for controlling vehicles in response to sensed environmental factors, and more specifically to an autonomous safety system or method for determining traffic characterization and performing margin management in real time.

BACKGROUND OF THE DISCLOSURE

[0002] Autonomous safety systems for vehicles are systems activated in response to one or more safety problems or abnormal events to avoid unwanted accidents. Such systems are typically used with one or more specific types of communications, such as Vehicle-to-Infrastructure (V2I), Vehicle-to- Vehicle (V2V), Vehicle-to-Pedestrian (V2P), Vehicle-to-Device (V2D), Vehicle-to-Grid (V2G), and Vehicle-to-Everything (V2X) communications, or the like. These safety systems can be activated by a human operator, or automatically by a computer system, or even mechanically.

[0003] For example, safety systems include vehicle dynamics control or active safety systems, such as yaw or roll stability control, adaptive cruise control, hill decent or ascent control, anti-lock brake system, electronic brake distribution, traction control system, suspension control system, steering control, drive-train control, engine control, and the like. However, conventional safety systems lack an ability to identify potential hazards in traffic conditions and/or severity of impending risks based on environmental characteristics related to the surroundings. Thus, an improved safety system is desired that evaluates the environmental characteristics and relevant factors such that potentially hazardous conditions, such as collisions or accidents, can be avoided or mitigated.

SUMMARY

[0004] In an embodiment of the present disclosure, a safety control system for a vehicle is provided, including a plurality of sensors associated with the vehicle configured to generate data related to at least one environmental characteristic; a system model generation unit configured to receive the data related to the at least one environmental characteristic, and to generate a system model based on the data, the system model representing a current environmental status of the vehicle based on the at least one environmental characteristic; a learning unit configured to receive the system model and learn the system model through a plurality of executions using the data related to the at least one environmental characteristic for performing a risk assessment of an environmental familiarity of the system model, the learning unit configured to generate an output value that determines at least one operating parameter of the vehicle such that the vehicle is operated based on the at least one operating parameter.

[0005] In one example, the learning unit performs the risk assessment using a dimensional look-up table having at least two input axes where each input axis is associated with the at least one environmental characteristic. In another example, the at least two input axes include a first input axis associated with a time of day, a second input axis associated with a location of the vehicle in latitude, and a third input axis associated with a location of the vehicle in longitude. In yet another example, the learning unit generates the output value using the dimensional look-up table. In still another example, the at least one environmental characteristic includes information related to at least one of: a road condition, a vehicle operator, a surrounding condition, a vehicle operating condition, an on-board data, a static horizon data, and a dynamic horizon data.

[0006] In a further example, the safety control system further includes an adjustment unit configured to adjust the output value based on one or more adjustment attributes associated with the at least one environmental characteristic. In one example, the adjustment unit detects a first adjustment attribute related to an environmental condition change, and modifies the output value based on the first adjustment attribute. In another example, the environmental condition change is at least one of: a temperature change, a traffic condition, and a weather condition. In yet another example, the adjustment unit detects a second adjustment attribute related to an uncertainty level of the output value, and modifies the output value based on the second adjustment attribute. In still another example, the uncertainty level is determined based on a trust score using an uncertainty table that provides a measure of an uncertainty of the output value. In still yet another example, the adjustment unit detects a third adjustment attribute related to an anomaly of at least one surrounding vehicle, and modifies the output value based on the third adjustment attribute. In a further example, the anomaly is determined based on a degree of deviation from an expected behavior of the at least one surrounding vehicle. In a yet further example, the adjustment unit detects a fourth adjustment attribute related to a risk factor of at least one surrounding vehicle, and modifies the output value based on the fourth adjustment attribute. In a still further example, the risk factor is determined based on identification of the at least one surrounding vehicle that have been identified as higher risk. In a still yet further example, the adjustment unit detects a fifth adjustment attribute related to sensor data degradation, and modifies the output value based on the fifth adjustment attribute. In one example, the sensor data degradation is determined based on a sensor data uncertainty level.

[0007] In another embodiment of the present disclosure, a method of mitigating a potentially hazardous condition for a vehicle is provided, and includes the steps of obtaining data related to at least one environmental characteristic using a plurality of sensors associated with the vehicle; receiving the data related to the at least one environmental characteristic and generating a system model based on the received data, the system model representing a current environmental status of the vehicle based on the at least one environmental characteristic; learning the system model through a plurality of executions using the data related to the at least one environmental characteristic; performing a risk assessment of an environmental familiarity of the system model; and generating an output value that determines at least one operating parameter of the vehicle such that the vehicle is operated based on the at least one operating parameter.

[0008] In one example, the method further includes using a dimensional look-up table having at least two input axes to perform the risk assessment, each input axis associated with the at least one environmental characteristic. In another example, the method further includes adjusting the output value based on one or more adjustment attributes associated with the at least one environmental characteristic. In yet another example, the method further includes overriding the output value by adjusting the at least one operating parameter related to the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above-mentioned and other features and advantages of this disclosure, and the manner of attaining them, will become more apparent and the present disclosure itself will be better understood by reference to the following description of embodiments of the present disclosure taken in conjunction with the accompanying drawings, wherein:

[0010] FIG. 1 illustrates an exemplary visualization of a safety control system featuring a central control unit;

[0011] FIG. 2 is a functional block diagram of the safety control system, featuring children units of the central control unit of FIG. 1;

[0012] FIG. 3 is a graphical representation of an exemplary dimensional look-up table or map used in the safety control system of FIG. 1;

[0013] FIG. 4 is a flow chart of an exemplary method of executing the safety control system of FIG. 1; and

[0014] FIGS. 5-9 are flow charts of exemplary methods of performing adjustments to an output value generated from the central control unit of FIG. 1 based on one or more adjustment attributes.

[0015] Corresponding reference characters indicate corresponding parts throughout the several views. Although the drawings represent embodiments of the present disclosure, the drawings are not necessarily to scale and certain features may be exaggerated in order to better illustrate and explain the present disclosure. The exemplifications set out herein illustrate an exemplary embodiment of the disclosure, in one form, and such exemplifications are not to be construed as limiting the scope of the disclosure in any manner.

DETAILED DESCRIPTION

[0016] Preferred embodiments of the present disclosure are described below by way of example only, with reference to the accompanying drawings. Further, the following description is merely exemplary in nature and is in no way intended to limit the disclosure, its application, or uses. As used herein, the term "unit" may refer to, be part of, or include an Application Specific Integrated Circuit (ASIC), an electronic circuit, a processor or microprocessor (shared, dedicated, or group) and/or memory (shared, dedicated, or group) that executes one or more software or firmware programs, a combinational logic circuit, and/or other suitable components that provide the described functionality. Thus, while this disclosure includes particular examples and arrangements of the units, the scope of the present safety control system should not be so limited since other modifications will become apparent to the skilled practitioner.

[0017] Referring now to FIG. 1, the present safety control system is generally designated 10, and is designed to provide an efficient way to control a vehicle 12, such as an on- or off-road automobile, using a central control unit (CCU) 14, such as an on-board computer having an engine control module. For example, during operation of vehicle 12, CCU 14 regulates an overall operation of safety control system 10 based on various environmental characteristics associated with vehicle 12, including information or data related to leading vehicles 12a or neighboring vehicles 16b-16d (collectively surrounding vehicles 16) and surrounding objects 18, e.g., road conditions and other obstacles. In general, CCU 14 monitors one or more environmental characteristics for controlling corresponding operating parameters of safety control system 10 of vehicle 12 via a network 20.

[0018] Any type of computer network having a collection of computers, servers, and other hardware interconnected by communication channels is contemplated, such as the Internet, Intranet, Ethernet, LAN, etc. In one embodiment, CCU 14 interfaces with network 20, such as a wired or wireless communication facility (e.g., a Wi-Fi access point), and performs control operations for the safety control system 10. Other similar networks known in the art are also contemplated. Based on one or more environmental characteristics, CCU 14 provides the operating parameters of safety control system 10, such as parameters related to powerplant, transmission, axle, air bags, brakes, steering, windshield wipers, external lighting, turn indicators, on-board OEM devices, telecom, warning indicators (e.g., audible, visual, or tactile information). For example, CCU 14 generates a minimum distance 22 (ΔάΜπΟ between vehicle 12 and one or more of neighboring vehicles 16b-16d or leading vehicles 16a and a maximum vehicle speed VMAX for vehicle 12.

[0019] In embodiments, the environmental characteristics are collected from one or more sensors 24 included in safety control system 10, such as surveying sensors (e.g., RADAR or LIDAR), cameras (e.g., visual spectrum or infrared spectrum), ultrasonics, yaw rate sensor, lateral acceleration sensor, steering angle sensor, grade sensor, vehicle sensors (e.g., mass or speed), accelerator pedal position sensor, ABS sensor, transmission gear state sensor, static map sensor, or the like, disposed inside or outside of vehicle 12 configured for measuring interior or exterior conditions relative to the environmental characteristics. However, it is also contemplated that the environmental characteristics is inputted into safety control system 10 by an operator or any other systems in vehicle 12. Other suitable methods of inputting the environmental characteristics are contemplated to suit different applications.

[0020] In embodiments, the environmental characteristics include road conditions, such as vehicle traffic, road grade, speed limits, lanes/exits/entry information, pedestrians, animals, or the like; vehicle operator information, such as age, mental state or alertness, aggressive or passive driving behavior, or the like; surrounding conditions, such as night or day, weather conditions (e.g., snow, rain, or fog), altitude, temperature, or the like; vehicle operating conditions, such as gross vehicle weight, on-board fuel amount, dynamic properties, or the like; on-board data from sensors, such as yaw rate sensor, lateral acceleration sensor, steering wheel sensor, wheel speed sensor, ABS sensor, RADAR sensor, ultrasonic sensor, LIDAR sensor, cameras (e.g., internal or external), accessories state sensor, transmission state sensor, engine state sensor, or the like; static horizon data, such as GPS data, road grade, road speed limits, number of lanes, intersection location, or the like; and dynamic horizon data via V2I, V2V, V2P, V2D, V2G, V2X communications, or the like. Other suitable configurations are also contemplated to suit different applications. [0021] Referring now to FIGS. 1 and 2, in one embodiment, during operation of safety control system 10, relevant information associated with environmental characteristics and operating parameters of vehicle 12 is displayed on an output device 26, such as an interactive display, accessible to the operator. CCU 14 manages interactions between the operator and safety control system 10 by way of a human machine interface (HMI), such as a keyboard, a touch sensitive pad or screen, a mouse, a trackball, a voice recognition system, and the like. Output device 26 (e.g., textual and graphical) is configured for receiving input data from the operator, CCU 14, or any other applications or systems associated with vehicle 12.

[0022] In one embodiment, the operator uses an input device, such as the HMI, to graphically or textually interact with safety control system 10. Associated data, information and/or parameters are generally received in CCU 14 and then transferred to output device 26 via a dedicated or shared communication system. Further, any collaborative other and third-party database reachable by CCU 14 can also be used for safety control system 10.

[0023] In one embodiment, safety control system 10 includes CCU 14 having an input acquisition unit 28, a system model generation unit 30, a learning unit 32, an adjustment unit 34, an alert and operating parameter generation unit 36, a storing unit 38, and a display unit 40. Although these sub-units 28, 30, 32, 34, 36, 38 are illustrated as children units subordinate of the parent unit CCU 14, each sub-unit can be operated as a separate unit from CCU 14, and other suitable combinations of sub-units are contemplated to suit different applications. One or more units or units can be selectively bundled as a key software model running on the processor having software as a service (SSaS) features.

[0024] All relevant information can be stored in a central database 42, e.g., as a non-transitory data storage device and/or a machine readable data storage medium carrying computer-executable instructions, for retrieval by CCU 14 and its children units. Also included in CCU 14 is an interface unit 42 for providing an interface between CCU 14, central database 42, network 20, and output device 26. Interface unit 42 controls operation of, for example, network 20, output device 26, and other related system devices, services, and applications. The other devices, services, and applications may include, but are not limited to, one or more software or hardware components, etc., related to CCU 14. Interface unit 42 also receives operating data or parameters from sensors 24 or other related systems, which are communicated to respective units, such as CCU 14, and its children units.

[0025] Input acquisition unit 28 is configured to receive one or more environmental characteristics via interface unit 42, and to transmit the received environmental characteristics to system model generation unit 30. Specifically, input acquisition unit 28 provides detailed information of interior and/or exterior environmental conditions, such as location and weather data, relative to vehicle 12 using one or more sensors 24. In general, as discussed in greater detail below, CCU 14 assesses operational status of vehicle 12 by evaluating the environmental characteristics.

[0026] System model generation unit 30 is configured to receive the environmental characteristics from input acquisition unit 28, and to examine the received environmental characteristics for controlling vehicle 12 based on a predetermined set of rules or techniques, as described below. In use, system model generation unit 30 generates a system model based on a predetermined analysis of the environmental characteristics, e.g., in communication with database 42. In embodiments, the system model represents a current environmental status of vehicle 12 determined based on the received environmental characteristics. An exemplary system model is shown as a dimensional look-up table or map in FIG. 3. Detailed descriptions are provided below in paragraphs relating to FIG. 3.

[0027] Learning unit 32 is configured to receive the system model from system model generation unit 30 and learn the system model through a plurality of executions using the environmental characteristics to perform a risk assessment of an environment familiarity of the system model. An exemplary learning method is based on learning an environment of vehicle 12 through repeated runs over a predetermined time period. In some embodiments, the learning method is in a form of closed circuit linear or non-linear model where unknown coefficients are approximately based on input data (e.g., Kalman filters, recursive least squares, optimized search, or the like). In other embodiments, the learning method is in a form of open circuit model, such as neural networks. In still other embodiments, the learning method is in a form of limited look-up tables or maps. Other suitable learning methods, such as dynamic structures that change based on uncertainty bounds, are also contemplated to suit the application.

[0028] In some embodiment, the Kalman filter is used to converge (or fuse) data into the dimensional look-up table to help establish a level of uncertainty associated with any given region in the dimensional look-up table. For example, new inputted data has less weight on adjusting the output values in the dimensional look-up table. Other suitable learning methods are also contemplated to suit different applications.

[0029] In embodiments, for a given system model of environment of vehicle 12, learning unit 32 measures an observability of the system model to perform the risk assessment of the environment familiarity of the system model. The observability represents a degree of how well the system model has learned the environment of vehicle 12. For example, due to a non-linear nature of environment measures, a measure of entropy based on an information theory is used to measure the observability of the system model. An exemplary information centric measure for observability H(x|y) can be defined by a function of a true value and a measured value, as provided by expression (1):

N M

H{* I y) = -∑∑/?(¾,^ )log2 P[X, I yj ) (1)

i=l j=l wherein x and y are random variables with joint probability distribution p(xi,yj), l≤i≤N, l≤j≤M. Expression (1) defines the conditional entropy of x given y.

[0030] An exemplary average amount of information about x contained in y can be defined in terms of a reduction in an uncertainty of x upon disclosure of y, as provided by expressions (2) and (3):

Figure imgf000011_0001
Figure imgf000012_0001
wherein In(x,y) reflects information of a current estimate of a dynamic parameter being estimated. In other words, an increasing certainty of a parameter estimate is reflected in an increasing value of In(x,y) associated with the parameter. Exemplary expressions (l)-(3) are further described in Compensating for Model Uncertainty in the Control of Cooperative Field Robots, Ph. D. thesis of Sujan, V. A, June 2002, Dept. of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, U. S. A, which is incorporated herein by reference.

[0031] Referring now to FIGS. 2 and 3, in some embodiments, learning unit 32 performs the risk assessment of the environment familiarity of the system model using a dimensional look-up table or map having at least two input axes where each input axis is associated with at least one environmental characteristic. For example, a set of latitude/longitude coordinates is used for the environmental characteristics. FIG. 3 shows an exemplary output space of the dimensional look-up table having a first input axis X associated with a time of day, a second input axis Y associated with a location of vehicle 12 in latitude, and a third input axis Z associated with a location of vehicle 12 in longitude. Learning unit 32 generates an output value for each axis that determines at least one operating parameter of vehicle 12 representing an optimal situational behavior of vehicle 12 based on the risk assessment using the dimensional look-up table. Other suitable axes related to one or more environmental characteristics are also contemplated to suit different applications. Also, other suitable methods, such as a closed network for weighted equation or a neural network, are contemplated, and other suitable variations of the axes are contemplated to suit the application. [0032] Adjustment unit 34 is configured to adjust or modify the output value based on one or more adjustment attributes associated with the environmental characteristics. Exemplary attributes include an environmental condition change, an uncertainty level of the output value, an anomaly of neighboring vehicles 16b-16d or leading vehicle 16a, one or more risk factors of neighboring vehicles 16b-16d or leading vehicle 16a, and sensor data degradation. In embodiments, adjustment unit 34 detects at least one environmental condition change and modifies the output value based on the detected environmental condition change.

[0033] For example, when the input axes are associated with longitude, latitude, and time of day, as temperature changes, the output value of each look-up table or map can be a maximum expected vehicle acceleration value, a maximum expected vehicle deceleration value, a minimum acceptable vehicle separation distance value, a maximum acceptable vehicle velocity, and the like. The expected values refer to values that are expected to be found in the system model, and the acceptable values refer to values that are deviated from the expected values by a predetermined amount. In some embodiments, the acceptable values include the expected values, and thus the expected values are a subset of the acceptable values. As another example, when the precipitation level increases at the same place and the same time of day, the maximum acceptable vehicle velocity is reduced to a lower value. Other suitable combinations of environmental characteristics are also contemplated to suit different applications.

[0034] In other embodiments, as vehicle 12 moves around, it looks at its current location (e.g., latitude and longitude) and time of day, and inputs such data to a single look-up table which then produces an output of the maximum expected vehicle acceleration. Additional look-up tables output the maximum expected vehicle deceleration, the minimum acceptable vehicle separation distance, and the maximum acceptable vehicle velocity, respectively. Thus, with these outputs, vehicle 12 is then able to constrain its movement, in the acceleration, velocity and separation distance space.

[0035] In some embodiments, another axis presenting a precipitation level, which can range from 0% to 100%, generates the maximum expected vehicle acceleration value, the maximum expected vehicle deceleration value, the minimum acceptable vehicle separation distance value, and the maximum acceptable vehicle velocity, using four dimensional look-up tables or maps.

[0036] Thus, FIG. 3 illustrates a simplified version of the output space of the dimensional look-up table depicting a hyper-dimensionality associated with one or more environmental characteristics. In this example, FIG. 3 represents a table having six inputs and one output. Thus, four such tables are needed to output the four outputs, namely the maximum expected vehicle acceleration value, the maximum expected vehicle deceleration value, the minimum acceptable vehicle separation distance value, and the maximum acceptable vehicle velocity.

[0037] For example, the dimensional look-up table initially includes no limiting output values for vehicle 12. In one embodiment, learning unit 32 learns the appropriate limiting output values as a function of location and time of vehicle operation. For example, if vehicle 12 is operated with CCU 14 in Indiana and used primarily in Indiana, then learning unit 32 only learns about roads in Indiana. However, if vehicle 12 is operated across Ohio, or any new locations, learning unit 32 continues to expand its learnings into those locations.

[0038] In some embodiments, learning unit 32 uses time and location data (e.g., six inputs as shown in FIG. 3 and the measured values for acceleration, deceleration, separation distance and vehicle velocity) and incorporates the data into the dimensional look-up table. For example, if the time and location data does not initially exist, learning unit 32 assumes that current measured values are the best values and inserts into the dimensional look-up table. However, after receiving the data for a predetermined period, a significant deviation can be detected based on historical data. When the significant deviation is detected, CCU 14 limits the behavior of vehicle 12 based on the output values created during previous cycles.

[0039] In other embodiments, learning unit 32 detects environmental changes and adjustment unit 34 adjusts the output values based on the environmental changes. For example, when vehicle 12 travels on a highway with a construction zone, learning unit 32 learns about the vehicle behavior in the construction zone (i.e., low velocity and acceleration output values). When the construction zone disappears, vehicle 12 desires to travel at a higher velocity. Initially, CCU 14 provides a lower maximum acceptable velocity due to the historical data traveling in the construction zone. After a predetermined time period, learning unit 32 detects the removal of the construction zone (i.e., environmental changes), and the output values are adjusted by adjustment unit 34. Thus, it is advantageous that CCU 14 provides best output values at any given time. In one embodiment, the output values are recommended values, but in another embodiment, the output values are powertrain commands. In some embodiments, adjustment unit 34 determines an uncertainty level of the output value. Each output value has an associated uncertainty level that represents how much the output value is trustworthy based on a trust score. For example, the trust score is determined using an uncertainty table or map that provides a measure of the uncertainty of the output value, and the uncertainty table is build based on the observability of the system model, as defined above. For example, as the uncertainty level decreases, the observability increases. An exemplary formulation of the output value VFINAL can be defined by a function of the output value VOUTPUT and the uncertainty level LEVELUNCERTAINTY, as provided by expression (4):

^ FINAL = VOUTPUT + K * LEVEL UNCERTAINTY (4) wherein K is a predetermined constant coefficient. For example, VOUTPUT denotes a maximum vehicle speed, and K denotes a predetermined vehicle speed.

[0040] Alert and operating parameter generation unit 36 is configured to generate an alert signal to inform the operator or other vehicle systems, and also configured to generate at least one operating parameter for vehicle 12 based on the output value. One or more warning messages are optionally sent by alert and operating parameter generation unit 36 to the operator or any other computing devices associated with vehicle 12. It is also contemplated that alert and operating parameter generation unit 36 provides an option to manually or automatically change or override the output value by adjusting one or more operating parameters related to vehicle 12. [0041] Storing unit 38 is configured to control and digitally store relevant information related to safety control system 10 in central database 42. More specifically, central database 42 includes operating data and parameters related to analysis data about the output value for the purposes of research, development, improvement of the comparative logic or methods and further investigations by the user or other related systems.

[0042] Display unit 40 is configured to interactively display an appropriate status or information message associated with safety control system 10 for illustration on output device 26. For example, an instance report related to each output value is generated by display unit 40, and also automatically transmitted to database 42, servers, or other systems, as desired.

[0043] Referring now to FIGS. 4-9, exemplary methods or processes of executing safety control system 10 are illustrated. Although the following steps are primarily described with respect to the embodiments of FIGS. 1-3, it should be understood that the steps within the methods can be modified and executed in a different order or sequence without altering the principles of the present disclosure.

[0044] Referring now to FIG. 4, the method begins at step 100. In step 102, input acquisition unit 28 receives at least one environmental characteristics, such as a location (e.g., GPS coordinates) of vehicle 12, time of day or year, weather conditions, an average traffic velocity of other vehicles 16, V2I or V2P information that provides a comparison point, or the like. In step 104, system model generation unit 30 examines the environmental characteristics based on a predetermined learning method that learns an environment of vehicle 12. In this example, an offline server collects a number of inputs from vehicles 12, 16 and develops an understanding of vehicle responses to a given input condition to create a correlation function for vehicle 12.

[0045] In step 106, system model generation unit 30 generates a system model based on a predetermined analysis of the environmental characteristics. In step 108, learning unit 32 receives the system model and learns the system model through a plurality of executions during a predetermined time period based on the environmental characteristics. Learning unit 32 performs a risk assessment of an environment familiarity of the system model to determine how well the system model has learned the environment of vehicle 12.

[0046] In step 110, learning unit 32 generates at least one output value that determines at least one operating parameter of vehicle 12 based on the risk assessment. Exemplary output values include deceleration and acceleration rates for vehicle 12, the minimum distance 22 (AdMiN) between vehicle 12 and one or more of neighboring vehicles 16b-16d or leading vehicle 16a and the maximum vehicle speed VMAX for vehicle 12. Other suitable output values are also contemplated to suit different applications. In step 112, if one or more adjustment attributes exist, adjustment unit 34 optionally adjusts or modifies the output value based on the adjustment attributes associated with vehicle 12. Depending on types of the adjustment attributes described above, additional steps shown in FIGS. 5-8 are performed. Further, it is also contemplated that alert and operating parameter generation unit 36 provides an option to manually or automatically change or override the adjustments or modifications by augmenting the output values associated with vehicle 12.

[0047] In step 114, when adjustment unit 34 determines that more adjustment attributes exist, control proceeds to step 112. Otherwise, control proceeds to step 116. In step 116, alert and operating parameter generation unit 32 generates an alert signal based on the output value to inform the operator or other vehicle system and generates at least one operating parameter for vehicle 12 based on the output value. Vehicle 12 is now operated based on the at least one operating parameter generated by safety control system 10. The method ends at step 118 which may include a return to step 102.

[0048] Referring now to FIG. 5, in step 200, when the adjustment attribute relates to an environmental condition change, e.g., weather conditions, and the environmental condition change is detected by adjustment unit 34, control proceeds to step 202. Otherwise, control proceeds to step 114. In step 202, the output value is changed based on the detected environmental condition change. For example, when vehicle 12 is operating on a rainy day, the maximum acceptable vehicle velocity is reduced to a lower value. [0049] Referring now to FIG. 6, in step 300, when the adjustment attribute relates to an uncertainty level of the output value, adjustment unit 34 determines the uncertainty level based on a trust score described above. In step 302, when the trust score representing the uncertainty level is lower than a predetermined threshold (i.e., less known due to lack of data), control proceeds to step 304. Otherwise, control proceeds to step 1 14. In some embodiments, the uncertainty level is quantified based on information theoretic formulations, such as Shannon' s Information Theory Measure of Entropy. Other suitable quantification techniques are also contemplated to suit different applications. In step 304, adjustment unit 34 adds a safety buffer to the output value, e.g., to the maximum vehicle speed VMAX and the minimum distance 22 (AdMiN)- For example, the maximum vehicle speed VMAX is reduced by a predetermined amount, or the minimum distance 22 (Δάι πΝ) is increased by a predetermined amount.

[0050] Referring now to FIG. 7, in step 400, when the adjustment attribute relates to an anomaly of neighboring vehicles 16b, 16c, 16d or leading vehicle 16a, adjustment unit 34 determines a degree of deviation from an expected behavior of neighboring vehicles 16b-16d or leading vehicle 16a. In step 402, when adjustment unit 34 determines that the degree of deviation is greater than a threshold value for a predetermined time period, control proceeds to step 404. Otherwise, control proceeds to step 1 14. In step 404, the output value is changed based on the degree of deviation. An exemplary formulation of the output value VFINAL can be defined by a function of the output value VMAX and the uncertainty level LEVELUNCERTAINTY, as provided by expression (5):

VFINAL = VMAX K * LEVELUNCERTAINTY— dt dAEXPECTED (5) wherein K is a predetermined constant coefficient. For example, VMAX denotes a maximum vehicle speed, K denotes a predetermined vehicle speed, dt denotes a delta time or time step, dA denotes a change in the expected acceleration rate, and dt*dA denotes a change in an expected velocity.

[0051] In some embodiments, a converged uncertainty increases based on a duration of deviation. An exemplary increased value 5NEW can be defined as provided by expression (6): ^NEW — ^ OLD +∑ dt +∑ dt · dAEXPECTED (6)

In some embodiment, future predictions of acceptable behaviors include the increased value 5NEW- In other embodiments, adjustment unit 34 monitors the behavior of one or more surrounding vehicles 16 in front of vehicle 12. Fewer surrounding vehicles 16 deviating from the expected behavior, fewer corrections to the increased value 5NEW occur. The monitoring of adjustment unit 34 can be achieved via V2V or direct sensing, and the monitoring results are used to guide the operator or to provide system override commands.

[0052] Referring now to FIG. 8, in step 500, when the adjustment attribute relates to risk factors of surrounding vehicles 16, adjustment unit 34 determines a risk factor associated with each surrounding vehicles 16 based on one or more measurements in the output space of the dimensional look-up table shown in FIG. 3. In some embodiments, additional measurements are considered, such as a maximum expected acceleration rate, a maximum expected deceleration rate, a minimum acceptable vehicle separation distance, a maximum acceptable vehicle velocity, a maximum acceptable vehicle lateral movement, and the like.

[0053] In embodiments, the risk factor is determined based on identification of surrounding vehicles 16 that have been identified as higher risk. For example, adjustment unit 34 determines the risk factor based on at least one of: a visual recognition, a vehicle information exchange, and an operator information exchange. Adjustment unit 34 quantifies the visual recognition based on a degree of deformation of a corresponding surrounding vehicle 16. For example, adjustment unit 34 calculates an amount of surface area of corresponding surrounding vehicles 16b, 16c, or 16d that is dissimilar as compared to a similar vehicle model. Thus, adjustment unit 34 calculates the risk factor commensurate with the degree of deformation of the corresponding surrounding vehicle.

[0054] In embodiments, adjustment unit 34 quantifies the vehicle information exchange based on data related to a degree of traffic violations associated with the corresponding surrounding vehicle 16. For example, via the V2X information exchange, adjustment unit 34 calculates a number of speeding or other traffic violations that the corresponding surrounding vehicle 16 committed during a predetermined time period. Thus, adjustment unit 34 calculates the risk factor commensurate with the degree of traffic violations. In some embodiments, during the calculation of the risk factor, a forgetting factor on past traffic violations is applied to remove old violation records. As such, recent historical traffic violations are weighted as higher risk than outdated violations.

[0055] In embodiments, adjustment unit 34 quantifies the operator information exchange based on information related to an operating behavior of the corresponding surrounding vehicle 16. For example, via the V2V information exchange, adjustment unit 34 determines whether the corresponding surrounding vehicle 16 is operating outside of the expected or acceptable values of the system model, and the risk factor is calculated commensurate with a number or degree of deviation from the expected or acceptable values of the system model. As such, safety control system 10 automatically provides in real-time an increased vehicle-to-vehicle buffer when a potentially aggressive or careless operator is identified.

[0056] In step 502, when adjustment unit 34 determines that the measurements are greater than a threshold value for a predetermined time period, control proceeds to step 504. Otherwise, control proceeds to step 114. In step 504, after the learning method has converged, if leading vehicle 16a is performing beyond modeled limits (e.g., the expected or acceptable values), a correction factor is added to the output value based on a duration and degree of deviation from the modeled limits. As another example, an increased vehicle to vehicle buffer is provided when leading vehicle 12a is compromised due to damaged vehicle components or operated by an impaired operator.

[0057] Referring now to FIG. 9, in step 600, when the adjustment attribute relates to sensor data degradation in vehicle 12, adjustment unit 34 determines a sensor data uncertainty level based on the sensor data degradation. For example, the sensor data uncertainty level is calculated based on an age of the corresponding sensor 24. In step 602, when adjustment unit 34 determines that the sensor data uncertainty level is greater than a threshold value, control proceeds to step 604. Otherwise, control proceeds to step 114. In step 604, the output value is adjusted or modified based on a degree of the sensor data uncertainty level. An exemplary formulation of the output value VFINAL can be defined by a function of the output value VMAX and an offset value (5OLD + NEW>, as provided by expression (7):

VFINAL = VMAX K * (<50iD + SNEW) (7) wherein K is a predetermined constant coefficient. For example, VMAX denotes a maximum vehicle speed, and K denotes a predetermined vehicle speed.

[0058] In some embodiments, the sensor data uncertainty level is determined based on a sensor data collection time. For example, when vehicle 12 is operating in daylight conditions but the output value is calculated using the sensor data collected in night conditions, the sensor data uncertainty level increases by a predetermined value. As with other adjustments described above, any adjustment attributes can be used to either guide the operator or provide system override commands for vehicle 12.

[0059] While the present disclosure has been described as having exemplary designs, the present disclosure can be further modified within the spirit and scope of this disclosure. This application is therefore intended to cover any variations, uses, or adaptations of the present disclosure using its general principles. Further, this application is intended to cover such departures from the present disclosure as come within known or customary practice in the art to which this present disclosure pertains and which fall within the limits of the appended claims.

[0060] Furthermore, the connecting lines shown in the various figures contained herein are intended to represent exemplary functional relationships and/or physical couplings between the various elements. It should be noted that many alternative or additional functional relationships or physical connections may be present in a practical system. However, the benefits, advantages, solutions to problems, and any elements that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as critical, required, or essential features or elements. The scope is accordingly to be limited by nothing other than the appended claims, in which reference to an element in the singular is not intended to mean "one and only one" unless explicitly so stated, but rather "one or more." [0061] Moreover, where a phrase similar to "at least one of A, B, or C" is used in the claims, it is intended that the phrase be interpreted to mean that A alone may be present in an embodiment, B alone may be present in an embodiment, C alone may be present in an embodiment, or that any combination of the elements A, B or C may be present in a single embodiment; for example, A and B, A and C, B and C, or A and B and C.

[0062] Systems, methods and apparatus are provided herein. In the detailed description herein, references to "one embodiment," "an embodiment," "an example embodiment," etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic with the benefit of this disclosure in connection with other embodiments whether or not explicitly described. After reading the description, it will be apparent to one skilled in the relevant art(s) how to implement the disclosure in alternative embodiments.

[0063] Furthermore, no element, component, or method step in the present disclosure is intended to be dedicated to the public regardless of whether the element, component, or method step is explicitly recited in the claims. No claim element herein is to be construed under the provisions of 35 U.S.C. § 112(f), unless the element is expressly recited using the phrase "means for." As used herein, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

Claims

CLAIMS WHAT IS CLAIMED IS:
1. A safety control system (10) for a vehicle (12), comprising:
a plurality of sensors (24) associated with the vehicle (12) configured to generate data related to at least one environmental characteristic;
a system model generation unit (30) configured to receive the data related to the at least one environmental characteristic, and to generate a system model based on the data, the system model representing a current environmental status of the vehicle (12) based on the at least one environmental characteristic;
a learning unit (32) configured to receive the system model and learn the system model through a plurality of executions using the data related to the at least one environmental characteristic for performing a risk assessment of an environmental familiarity of the system model, the learning unit (32) configured to generate an output value that determines at least one operating parameter of the vehicle (12) such that the vehicle (12) is operated based on the at least one operating parameter.
2. The safety control system (10) of claim 1, wherein the learning unit (32) performs the risk assessment using a dimensional look-up table having at least two input axes where each input axis is associated with the at least one environmental characteristic.
3. The safety control system (10) of claim 2, wherein the at least two input axes include a first input axis associated with a time of day, a second input axis associated with a location of the vehicle (12) in latitude, and a third input axis associated with a location of the vehicle (12) in longitude.
4. The safety control system (10) of claim 2, wherein the learning unit (32) generates the output value using the dimensional look-up table.
5. The safety control system (10) of claim 1, wherein the at least one environmental characteristic includes information related to at least one of: a road condition, a vehicle operator, a surrounding condition, a vehicle operating condition, an on-board data, a static horizon data, and a dynamic horizon data.
6. The safety control system (10) of claim 1, further comprising an adjustment unit (34) configured to adjust the output value based on one or more adjustment attributes associated with the at least one environmental characteristic.
7. The safety control system (10) of claim 6, wherein the adjustment unit (34) detects a first adjustment attribute related to an environmental condition change, and modifies the output value based on the first adjustment attribute.
8. The safety control system (10) of claim 7, wherein the environmental condition change is at least one of: a temperature change, a traffic condition, and a weather condition.
9. The safety control system (10) of claim 6, wherein the adjustment unit (34) detects a second adjustment attribute related to an uncertainty level of the output value, and modifies the output value based on the second adjustment attribute.
10. The safety control system (10) of claim 9, wherein the uncertainty level is determined based on a trust score using an uncertainty table that provides a measure of an uncertainty of the output value.
11. The safety control system (10) of claim 6, wherein the adjustment unit (34) detects a third adjustment attribute related to an anomaly of at least one surrounding vehicle (16), and modifies the output value based on the third adjustment attribute.
12. The safety control system (10) of claim 11, wherein the anomaly is determined based on a degree of deviation from an expected behavior of the at least one surrounding vehicle (16).
13. The safety control system (10) of claim 6, wherein the adjustment unit (34) detects a fourth adjustment attribute related to a risk factor of at least one surrounding vehicle (16), and modifies the output value based on the fourth adjustment attribute.
14. The safety control system (10) of claim 13, wherein the risk factor is determined based on identification of the at least one surrounding vehicle (16) that have been identified as higher risk.
15. The safety control system (10) of claim 6, wherein the adjustment unit (34) detects a fifth adjustment attribute related to sensor data degradation, and modifies the output value based on the fifth adjustment attribute.
16. The safety control system (10) of claim 15, wherein the sensor data degradation is determined based on a sensor data uncertainty level.
17. A method of mitigating a potentially hazardous condition for a vehicle (12), comprising:
obtaining data related to at least one environmental characteristic using a plurality of sensors (24) associated with the vehicle (12);
receiving the data related to the at least one environmental characteristic and generating a system model based on the received data, the system model representing a current environmental status of the vehicle (12) based on the at least one environmental characteristic;
learning the system model through a plurality of executions using the data related to the at least one environmental characteristic;
performing a risk assessment of an environmental familiarity of the system model; and
generating an output value that determines at least one operating parameter of the vehicle (12) such that the vehicle (12) is operated based on the at least one operating parameter.
18. The method of claim 17, further comprising using a dimensional look-up table having at least two input axes to perform the risk assessment, each input axis associated with the at least one environmental characteristic.
19. The method of claim 17, further comprising adjusting the output value based on one or more adjustment attributes associated with the at least one environmental characteristic.
20. The method of claim 17, further comprising overriding the output value by adjusting the at least one operating parameter related to the vehicle (12).
PCT/US2017/055751 2017-10-09 2017-10-09 Autonomous safety systems and methods for vehicles WO2019074478A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/US2017/055751 WO2019074478A1 (en) 2017-10-09 2017-10-09 Autonomous safety systems and methods for vehicles

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2017/055751 WO2019074478A1 (en) 2017-10-09 2017-10-09 Autonomous safety systems and methods for vehicles

Publications (1)

Publication Number Publication Date
WO2019074478A1 true WO2019074478A1 (en) 2019-04-18

Family

ID=66101590

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2017/055751 WO2019074478A1 (en) 2017-10-09 2017-10-09 Autonomous safety systems and methods for vehicles

Country Status (1)

Country Link
WO (1) WO2019074478A1 (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6269307B1 (en) * 1998-08-06 2001-07-31 Honda Giken Kogyo Kabushiki Kaisha Travel safety system for vehicle
US7839292B2 (en) * 2007-04-11 2010-11-23 Nec Laboratories America, Inc. Real-time driving danger level prediction
US8471726B2 (en) * 2010-03-03 2013-06-25 Volvo Car Corporation System and method for collision warning
US9176500B1 (en) * 2012-05-14 2015-11-03 Google Inc. Consideration of risks in active sensing for an autonomous vehicle
US20160282874A1 (en) * 2013-11-08 2016-09-29 Hitachi, Ltd. Autonomous Driving Vehicle and Autonomous Driving System
US20170113702A1 (en) * 2015-10-26 2017-04-27 Active Knowledge Ltd. Warning a vehicle occupant before an intense movement
US20170120904A1 (en) * 2015-11-04 2017-05-04 Zoox, Inc. Robotic vehicle active safety systems and methods

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6269307B1 (en) * 1998-08-06 2001-07-31 Honda Giken Kogyo Kabushiki Kaisha Travel safety system for vehicle
US7839292B2 (en) * 2007-04-11 2010-11-23 Nec Laboratories America, Inc. Real-time driving danger level prediction
US8471726B2 (en) * 2010-03-03 2013-06-25 Volvo Car Corporation System and method for collision warning
US9176500B1 (en) * 2012-05-14 2015-11-03 Google Inc. Consideration of risks in active sensing for an autonomous vehicle
US20160282874A1 (en) * 2013-11-08 2016-09-29 Hitachi, Ltd. Autonomous Driving Vehicle and Autonomous Driving System
US20170113702A1 (en) * 2015-10-26 2017-04-27 Active Knowledge Ltd. Warning a vehicle occupant before an intense movement
US20170120904A1 (en) * 2015-11-04 2017-05-04 Zoox, Inc. Robotic vehicle active safety systems and methods

Similar Documents

Publication Publication Date Title
Moon et al. Design, tuning, and evaluation of a full-range adaptive cruise control system with collision avoidance
Baskar et al. Traffic control and intelligent vehicle highway systems: a survey
US7418372B2 (en) Model predictive control apparatus
US9575490B2 (en) Mapping active and inactive construction zones for autonomous driving
US8140241B2 (en) Driving action estimating device, driving support device, vehicle evaluating system, driver model creating device, and driving action determining device
US9081651B2 (en) Route navigation with optimal speed profile
Naranjo et al. Using fuzzy logic in automated vehicle control
US7974748B2 (en) Driver assistance system with vehicle states, environment and driver intention
CN102076541B (en) Path generation algorithm for automated lane centering and lane changing control system
US9552735B2 (en) Autonomous vehicle identification
US8949016B1 (en) Systems and methods for determining whether a driving environment has changed
US8437890B2 (en) Integrated framework for vehicle operator assistance based on a trajectory prediction and threat assessment
US9663117B2 (en) Engaging and disengaging for autonomous driving
CN102216119B (en) The method of the chaufeur of vehicle and prompting vehicle
US10379533B2 (en) System and method for autonomous vehicle fleet routing
McCall et al. Driver behavior and situation aware brake assistance for intelligent vehicles
US9460622B1 (en) Approach for estimating the geometry of roads and lanes by using vehicle trajectories
JP5690322B2 (en) A vehicle with a computer that monitors and predicts objects participating in traffic
US9517771B2 (en) Autonomous vehicle modes
US9811091B2 (en) Modifying behavior of autonomous vehicles based on sensor blind spots and limitations
US9754501B2 (en) Personalized driving ranking and alerting
Desjardins et al. Cooperative adaptive cruise control: A reinforcement learning approach
DE112010005774T5 (en) Systems and methods for scheduling driver interface tasks based on driver workload
US10179589B2 (en) System and method for optimizing fuel economy using predictive environment and driver behavior information
US20160026182A1 (en) Personalized Driving of Autonomously Driven Vehicles

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17928131

Country of ref document: EP

Kind code of ref document: A1

DPE1 Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101)