This disclosure is related to vehicle control.
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
Vehicle control describes methodologies wherein operation of a vehicle or of a system within a vehicle is monitored and computerized control is applied to modify the operation of the vehicle. Exemplary vehicle control strategies monitor vehicle dynamics or the current or expected forces and moments acting upon the vehicle and apply control methods based upon the forces and moments to maintain desired operation of the vehicle. For example, when a vehicle is stopping quickly, forces and moments can cause the body of the vehicle to rotate forward, compressing the front suspension. Vehicle control can counter such monitored or anticipated rotation of the body of the vehicle by energizing actuators located at the front corners of the vehicle to lift up on the vehicle body and stop the front suspension from compressing. In another example, in a braking maneuver, forces and moments acting upon the vehicle and braking forces being applied to the wheels can be used to monitor or predict slippage between a wheel and the road. Actuators at the brakes of the vehicle can be used to redistribute braking force among the wheels to avoid slippage. In another example, in a turning maneuver, forces and moments acting upon the vehicle and a steering angle can be used to monitor or predict an over-steer or an under-steer condition, and an actuator acting upon the steering system can correct the steering. In another example, forces acting upon the vehicle and a current torque being applied to the drivetrain and the wheels can be used to monitor or predict slippage between the wheels and the road, and an actuator can be used to modify the torque being transmitted to the wheels through various methods known in the art.
A number of systems can be monitored or affected through vehicle control. Examples include braking, steering, driveline, and suspension control. According to an exemplary method, driver commands and/or autonomous driving system commands are integrated to determine desired vehicle dynamic or kinematic states, a vehicle dynamics module inputs the desired vehicle dynamic or kinematic states and outputs a desired vehicle force and/or moment, and an actuator supervisory control module generates commands in one or more vehicle systems to effect the desired vehicle force and or moment. Integration of the commands can include feedback regarding current or resultant vehicle dynamic or kinematic states, and the actuator supervisory control module can include feedback of current or resultant actuator states.
A method to control a vehicle including control of regenerative brakes and friction brakes includes monitoring a desired corner force and moment distribution, monitoring real-time actuator constraints including a braking torque limit of each of the regenerative brake, determining a regenerative braking torque for each of the regenerative brakes based upon the desired corner force and moment distribution and the real-time actuator constraints, determining a friction braking torque for each of the friction brakes based upon the desired corner force and moment distribution and the determined regenerative braking torque for each of the regenerative brakes, and controlling the vehicle based upon the determined regenerative braking torques and the determined friction braking torques.
One or more embodiments will now be described, by way of example, with reference to the accompanying drawings, in which:
Referring now to the drawings, wherein the showings are for the purpose of illustrating certain exemplary embodiments only and not for the purpose of limiting the same,
As described above, methods are known to utilize actuators to modify the operation of various vehicle systems for purposes of vehicle control. For braking systems, electronic stability control systems (ESC) and antilock braking systems are systems that modulate braking actuation based upon inputs. Vehicle control inputs can be utilized to change the operation of the ESC or antilock braking systems to affect actuator control of braking. ESC, antilock braking systems, or other braking control systems can use hydraulic actuators with pumps and valves, self-pressurized systems, electric motors, or electric calipers to affect actuator control of braking. The disclosure is not intended to be limited to the particular exemplary embodiments described herein.
Commonly assigned U.S. Pat. No. 7,472,006 is entitled “VEHICLE DYNAMICS CONTROL STRUCTURE” describes a method to implement vehicle control based upon coordinating individual vehicle systems and is hereby incorporated by reference.
Method 100 provides for integrated control of various vehicle systems for vehicle control. Actuator supervisory control module 130 applies methods and programmed responses to the desired vehicle force and moment 132 to the various vehicle systems commanded by module 140, 145 and 150. However, these methods and programmed responses are feature-based responses, for example, a desired moment is identified, and a response including modification of a steering angle is generated. Such a response is still an isolated function for that vehicle system.
A corner dynamics control method is applied to a vehicle control method, determining a desired corner force and moment distribution for every wheel of the vehicle. By distributing a desired vehicle force and moment as a desired corner force and moment to each of the corners of the vehicle, a response of the various systems affecting a particular corner can be selectively generated based upon the desired force and moment for that corner. Such a method implements vehicle control as a net effect of all of the vehicle systems controlled rather than controlling the individual system piecemeal. Further, by distributing the desired force and moment to the four corner of the vehicle, real-time corner constraints or constraints describing an ability of each corner to contribute to vehicle control can be applied as part of the distribution, ensuring that the desired corner force and moment distribution is within a desired range for each corner and not in a near-limit condition. In this way, integration of various vehicle systems can be achieved based upon the vehicle response to the entirety of the systems being controlled. Further, by unifying control based upon desired forces and moments acting upon the vehicle, the control parameters can be standardized or customizable for vehicle configuration and driver preferences.
Vehicle control can take a number of embodiments. As described in association with
Actuator anomalies can reduce the ability of a vehicle control system to react to vehicle operation. For example, if an actuator is non-functional or does not react to commands generated by the vehicle control system, benefits from the vehicle control system can be negated or adverse effects to drivability can be apparent. Based upon a system utilizing a corner dynamics control module as described herein monitoring actuator function and determining an actuator to be non-functional, distribution of force and moment to the various corners can be adjusted to compensate for the known actuator anomaly.
According to one exemplary vehicle configuration, a vehicle can utilize, in parallel or selectively, both regenerative braking and friction braking to reduce the speed of the vehicle. In the exemplary configuration depicted in
Vehicle control methods based upon a desired force and moment acting upon the vehicle center of gravity of a vehicle may include open-loop table-lookup values, for example calibrated through intensive tuning for various driving conditions, or iterative calculations such as a least square method to achieve optimal performance, each individually controlling vehicle systems based upon the desired force and moment. However, these methods, including calibrated responses for the various vehicle systems to the desired force and moment input react based upon the behavior of that particular system or calibrated response. As a result, such responses include optimized minimum or maximum values for that given system that are local to that system. However, viewing the vehicle control response as a whole including all of the vehicle systems controlled, such local minimum or maximum values for a particular system or for a specific calibration can be different from an optimized vehicle-wide solution including responses based upon desired corner force and moment distribution. Further, iterative calculations for vehicle control can include a delay while the iterative calculations are resolved to a solution. Under dynamic or urgent driving conditions, such a delay can be undesirable for vehicle control methods. Instead, a closed-loop calculation can be preferable for vehicle control wherein a determinative answer can be solved for a given set of inputs.
The methods of the present disclosure include closed-loop calculations based upon finding an optimal solution in a closed-form that helps to avoid iteration and reduces the need for intensive tuning the controller. Vehicle longitudinal and lateral forces and yaw moment at the center of gravity can be determined based upon tire forces. Ignoring self-aligning tire moments, vehicle dynamic equations describing longitudinal force upon the vehicle, Fx, lateral force upon the vehicle, Fy, and moment acting on a horizontal plane upon the vehicle, Gz, can be expressed as follows.
Fx=Fx(Fx1,Fy1,Fx2,Fy2,Fx3,Fy3,Fx4,Fy4) [1]
Fy=Fy(Fx1,Fy1,Fx2,Fye,Fx3,Fy3,Fx4,Fy4) [2]
Gz=Gz(Fx1,Fy1,Fx2,Fy2,Fx3,Fy3,Fx4,Fy4) [3]
Subscripts indicate the forces acting upon each of the individual tires. Equations 1-3 describing the forces acting at the vehicle's center of gravity as resultants of those applied from the road surface to the four tires can be expressed in the following linear form.
The term AF3x8 describes a matrix that depends upon the vehicle geometry and road wheel angles as follows:
The term T describes a track width or a width defined by the distance between the centerlines of tires on a left and right side of the vehicle at the ground level. The term δi describes the steer angle for each of the wheels. The terms a and b represent the distances between the center of gravity and the front and rear axles, respectively.
The linear form of the vehicle dynamic equations with respect to the corner forces enables use of an exemplary standard optimization methodology as follows.
The terms f and δf describe vectors of tire forces and corner force control adjustments, respectively. The terms Fx*, Fy* and Gz* describe target or desired longitudinal force, lateral force, and the rotational moment, respectively, determined by the driver command interpreter. Optimization of the vehicle dynamics can be accomplished by minimizing an error vector, E. The error vector is introduced in Equation 6 as follows.
Similarly, a tire forces-center of gravity forces transition matrix, AF, can be described as follows.
Equation 6 therefore yields the following equation.
The terms (E−AFδf) can collectively be described as a closed loop tracking error. The term δf is the tire force variation to be determined. The following expression describes a friction ellipse physical constraint on tire force distribution due to tire characteristics.
Fxi,max and Fyi,max describe maximum forces that can be exerted by the tires in longitudinal and lateral directions, respectively, and can be described as follows.
Fxi,max=μxFzi [11]
Fyi,max=μyFzi [12]
Equation 9 and 10 together can be utilized to optimize corner force distribution by minimizing the following cost function.
Of the components of Equation 13, the following expression describes a center of gravity force error component or error expression to be minimized:
Of the components of Equation 13, the following expression describes a constraint on tire force control variation or control energy component to be minimized:
Of the components of Equation 13, the following expression describes a fast growing tire force near the friction ellipse of any tire or an expression describing tire force reserve component to be maximized:
In order to minimize the number of parameters, all the weighting matrixes can be represented in the diagonal form, as follows.
It is essential that Equation 17 includes no tire force control variations as compared to Equation 10. As a result, the cost function from Equation 13 has the quadratic form with respect to tire force control adjustments.
Using Equation 13 and assuming as a necessary condition for finding a minimum that
a closed form solution for determining optimized tire force distribution or a real-time closed form dynamics optimization solution is obtained in the form as follows.
δf=[Wf+Wdf+(AFTWE)AF]−1[AFT(WEE)−Wff]det[Wf+Wdf+(AFTWE)AF]≠0 [21]
In this way, a system including exemplary corner dynamics control module 220 can generate the exemplary desired corner force and moment distribution 232 signal described above, which can subsequently be used in vehicle control, as described above.
The desired corner force and moment distribution is based upon inputs including driver inputs, can be based upon or incorporate a desired braking command, for example, describing a desired braking torque. Q desired describes a total torque value for the vehicle that the four corners of the vehicle or the braking systems of the wheels of the vehicle are commanded to generate. However, these commands must be distributed as corner specific and brake system specific commands. As described above, a desired corner force and moment distribution can be determined describing a force and moment desired of each corner of the vehicle. Based upon the methods described above, terms Fx1, Fx2, Fx3, Fx4 can be determined, describing desired longitudinal forces distributed to each corner of the vehicle. A desired braking torque for each wheel or corner of the vehicle, Qidesired, can be determined as a function of Fxi, as expressed by the following.
Reffective describes an effective radius of the wheel and can be used to describe not only an actual radius of the wheel, describing a relationship of torque to axial force at the circumference of a wheel, but also the effects of other factors upon a force resulting from a torque applied.
An exemplary method to control braking in a vehicle includes determining and minimizing force and moment errors. Corner-based actuator torque equations can be expressed as follows.
Qim describes regenerative braking torque applied by motors or regenerative dampers at each of the four wheels. Qib describes friction braking torque applied to each of the front wheels. The resultant torque at each of the front corners i=1, 2 can be described as a sum of Qim and Qib. λi and εi are control variables. λ describes a friction braking distribution ratio, wherein at zero, all braking is performed by regenerative braking and, at one, all braking is performed by friction braking. ε provides necessary increase of friction torques on the front wheels when regenerative braking torques on the rear wheels are insufficient to meet the desired longitudinal force and yaw moment applied to the vehicle center of gravity. A measured or actual center of gravity force and moment, FCGactual, can be described as follows:
AQ describes a corner to center of gravity transformation matrix and Qi (i=1,2,3,4) describes the resultant torque on each of the four wheels. By determining a desired center of gravity force and moment, FCGdesired, and comparing it to FCGactual, a center of gravity force error can be determined and minimized. Such a minimizing of force errors can be described by the following expression
under the following constraint conditions.
Qim
Qib
W(t)=Σi=14ωiQim(t)=−Σi=14ωi(1−λi)Qidesired(t)=Nλ−Σi=14ωiQidesired(t)≦Wmax(t) [28]
Qm
H(x) describes the Heaviside unit-step function, which is zero at x<0 and one at x≧0. As depicted in
λ can be optimized for every corner of the vehicle, such that each corner utilizes regenerative braking at that corner until Qim
Qim=(1−λi)Qidesired [30]
For wheels including a plurality of braking systems, λi can be utilized to distribute torque between regenerative braking and friction braking at that corner. For wheels only including regenerative braking, λi can be utilized to distribute torque to other wheels of the vehicle, for example, utilizing regenerative braking in the wheel without friction braking to provide a desired braking torque for the wheel up to a braking torque limit for that wheel and distributing a remaining portion of the desired braking torque for the wheel to another wheel of the vehicle that includes a friction brake. In an exemplary front wheel of the configuration depicted in
Q1b=(λ1+ε1)Q1desired [31]
ε1 is a compensation parameter that can be used to increase Q1b based upon factors including λ values for the wheels that do not have friction braking. An exemplary expression for ε corresponding to the exemplary configuration of
Equation 32 and similar equation for another front wheel together represent an explicit solution of optimization problem depicted in Equation 25. In the exemplary embodiment of
Distribution of braking torque can include other factors affecting braking. Another exemplary factor takes into account total power management for the vehicle. According to one embodiment, λi values as determined by Equation 28 can be adjusted according to the following.
The term −ζini modifies Equation 29 in order to reduce regenerative braking as the total power W(t) described by Equation 28 is reaching its maximal value as shown in
The region above plane 450 describes prohibited half-space, while the region below plane 450 describes allowed half space. Vector 460 represents the direction of a quickest escape from the constraint based upon the meaning of the normal vector 470. Line 480 represents a weighted escape. An exemplary weighted escape may not coincide with vector 460 and can be described by the following set of components: −{ζ1n1, ζ2, ζ3n3, ζ4n4}, where the weights are described by the following.
ζi=ζ(W−Wmax;rxiryi) [35]
wherein, ζ describes a function illustrated by
rxi=rxi(|Qidesired−Qiactual|) [36]
and
ryi=ryi(|Qidesired−Qiactual|) [37]
describe parameters of the function that may depend on errors between desired and actual total torques on each of the wheels. The parameters rxi and ryi are to modify the scales along the horizontal and vertical axes, respectively, and may affect escape rates from plane 450 in
The above methods describe exemplary equations and terms based upon the configuration of
The above methods describe methods of vehicle control that can be used to control vehicle systems. Vehicle systems can take many embodiments. For example, a powertrain is depicted in
The above embodiments of the vehicle control method primarily treat vehicle forces and moments as existing in a two-dimensional plane close to or coincident with the horizontal ground plane. However, it will be appreciated that a third dimension can be utilized in vehicle control methods. By addition of similar modules and constraints in a third-dimension, it will be appreciated that vehicle control methods can be described that include three dimensions of control including, for example, forward and rearward roll of the vehicle body in relation to the chassis of the vehicle and corresponding corner force and moments to maintain control over the vehicle. A roll control system is known to control dynamics of a vehicle body in relation to a vehicle chassis. Such control can be based upon information in either three dimensions or information in two dimensions and estimating an effect of that two dimensional data to roll.
The above methods describe utilizing a desired corner force and moment distribution describing desired braking torques for wheels of the vehicles to distribute braking torques to various braking devices at the different wheels.
Control module, module, controller, control unit, processor and similar terms mean any suitable one or various combinations of one or more of Application Specific Integrated Circuit(s) (ASIC), electronic circuit(s), central processing unit(s) (preferably microprocessor(s)) and associated memory and storage (read only, programmable read only, random access, hard drive, etc.) executing one or more software or firmware programs, combinational logic circuit(s), input/output circuit(s) and devices, appropriate signal conditioning and buffer circuitry, and other suitable components to provide the described functionality. The control module has a set of control algorithms, including resident software program instructions and calibrations stored in memory and executed to provide the desired functions. The algorithms are preferably executed during preset loop cycles. Algorithms are executed, such as by a central processing unit, and are operable to monitor inputs from sensing devices and other networked control modules, and execute control and diagnostic routines to control operation of actuators. Loop cycles may be executed at regular intervals, for example each 3.125, 6.25, 12.5, 25 and 100 milliseconds during ongoing engine and vehicle operation. Alternatively, algorithms may be executed in response to occurrence of an event.
The disclosure has described certain preferred embodiments and modifications thereto. Further modifications and alterations may occur to others upon reading and understanding the specification. Therefore, it is intended that the disclosure not be limited to the particular embodiment(s) disclosed as the best mode contemplated for carrying out this disclosure, but that the disclosure will include all embodiments falling within the scope of the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5358317 | Cikanek | Oct 1994 | A |
6222334 | Tamagawa et al. | Apr 2001 | B1 |
6318487 | Yanase et al. | Nov 2001 | B2 |
6325470 | Schneider | Dec 2001 | B1 |
6454364 | Niwa et al. | Sep 2002 | B1 |
6488344 | Huls et al. | Dec 2002 | B2 |
6837552 | Reuter et al. | Jan 2005 | B2 |
7427849 | Kaneko et al. | Sep 2008 | B2 |
7472006 | Turski et al. | Dec 2008 | B2 |
20060155451 | Kuwahara et al. | Jul 2006 | A1 |
20060196712 | Toyota et al. | Sep 2006 | A1 |
20070004553 | Oikawa et al. | Jan 2007 | A1 |
20070021875 | Naik et al. | Jan 2007 | A1 |
20080066976 | Yamada | Mar 2008 | A1 |
20080140283 | Kuwahara et al. | Jun 2008 | A1 |
20100268420 | Yasui et al. | Oct 2010 | A1 |
Number | Date | Country |
---|---|---|
102006055766 | Oct 2007 | DE |
2008301590 | Dec 2008 | JP |
2011 101 515 | May 2011 | JP |
Entry |
---|
English translation of DE-102006055766 A1. |
U.S. Appl. No. 12/845,127, filed Jul. 28, 2010, Chen et al. |
Svendenius, J., “PhD Thesis: Tire Modeling and Friction Estimation, Department of Automatic Control”, Lund University, Lund, Apr. 2007. |
He et al., “Experimental Evaluation of Dynamic Force Distribution Method for EV Motion Control”, IEEE Ref. No. 1-4244-0844-X/07, pp. 682-687, 2007. |
Hamada et al., “Development of Vehicle Dynamics Integrated Management”, SAE Paper No. 2006-01-0922, Apr. 3-6, 2006. |
Brach et al., “Modeling Combined Braking and Steering Tire Forces”, SAE Paper No. 200-01-0357, Mar. 6-9, 2000. |
Nicolas et al., “Predicting Directional Behavior of Tractor Semitrailers When Wheel Anti-Skid Brake Systems Are Used”, Paper No. 72 WA/Aut-16, ADME Winter Annual Meeting, Nov. 26-30, 1972. |
Pottinger et al., “Effectiveness of the Slip Circle, “Combinator”, Model for Combined Tire Cornering and Braking Forces When Applied to a Range of Tires”, SAE Paper 982747, Nov. 16-18, 1998, Warrendale, PA 15096. |
Number | Date | Country | |
---|---|---|---|
20120055744 A1 | Mar 2012 | US |