CN114103926A - Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment - Google Patents

Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment Download PDF

Info

Publication number
CN114103926A
CN114103926A CN202111495143.0A CN202111495143A CN114103926A CN 114103926 A CN114103926 A CN 114103926A CN 202111495143 A CN202111495143 A CN 202111495143A CN 114103926 A CN114103926 A CN 114103926A
Authority
CN
China
Prior art keywords
vehicle
coordinate system
speed
hybrid
tracked vehicle
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202111495143.0A
Other languages
Chinese (zh)
Inventor
项昌乐
郭凌雄
刘洋
刘辉
韩立金
徐丽丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
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 Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN202111495143.0A priority Critical patent/CN114103926A/en
Publication of CN114103926A publication Critical patent/CN114103926A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W20/00—Control systems specially adapted for hybrid vehicles
    • B60W20/10—Controlling the power contribution of each of the prime movers to meet required power demand
    • B60W20/15—Control strategies specially adapted for achieving a particular effect
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0002—Automatic control, details of type of controller or control system architecture
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2300/00—Indexing codes relating to the type of vehicle
    • B60W2300/44—Tracked vehicles
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/80—Technologies aiming to reduce greenhouse gasses emissions common to all road transportation technologies
    • Y02T10/84—Data processing systems or methods, management, administration

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)

Abstract

The invention belongs to the technical field of vehicle collaborative optimization driving, and discloses a hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment, which comprise the following steps: designing a coordinate transformation method to convert the GPS coordinates of the original set track into the coordinates of a vehicle local coordinate system so as to facilitate the application of a subsequent algorithm; constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle; by utilizing the reinforcement learning method, tracking errors and energy consumption are fully considered, so that the possible behaviors of slipping, side shifting, oversteering and the like in the path following process of the vehicle are reduced, and the speed and energy collaborative optimization in the path following process of the hybrid tracked vehicle is realized. The invention improves the behaviors of slipping, oversteering and the like of the vehicle in the path following process by optimizing the running speed sequence of the tracks on the two sides of the tracked vehicle, thereby realizing the cooperative optimization of the speed and the energy of the tracked vehicle.

Description

Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment
Technical Field
The invention belongs to the technical field of vehicle collaborative optimization driving, and particularly relates to a hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment.
Background
At present, the energy management technology combined with speed planning is continuously developed, and the energy utilization rate of the vehicle is improved by optimizing the running speed curve of the hybrid vehicle, namely the ecological driving technology oriented to the hybrid vehicle is tried, so that the energy consumption of the vehicle can be obviously reduced. At this stage, the related art generally adopts a hierarchical optimization method, and the controller is generally hierarchically divided into two layers: 1) a vehicle layer (also referred to as an outer layer) in which the target vehicle autonomously plans a vehicle speed according to a specific driving environment; 2) and the dynamic power assembly is layered (inner layer), and optimal energy distribution among the power modules is realized based on the planned speed curve. Although the calculation cost in the control process can be reduced by the hierarchical optimization, the coupling relation between the vehicle speed and the torque distribution is not well processed in the optimization process, and the cooperative optimization of the speed planning and the torque distribution cannot be realized. And the related research of the United states department of energy shows that: more fuel savings can be expected if both vehicle-level and powertrain-level control can be optimized. Therefore, how to realize the coordination of speed planning and energy management is a great difficulty to be considered urgently.
Currently, research on how to realize cooperative optimization of speed planning and energy management to further promote energy saving of hybrid vehicles is still few, and research objects mainly focus on the aspect of civil wheeled vehicles, and a cooperative optimization problem facing tracked vehicles is urgently needed to be solved. The steering process of the civil wheeled vehicle based on Ackerman steering is generally regarded as no energy loss, so that the conventional scheme decouples the speed and energy collaborative optimization problem into two problems of longitudinal speed planning and steering tracking control, thereby simplifying the problem difficulty and realizing effective application. However, unlike civil wheeled vehicles based on ackerman steering, tracked vehicles employ differential steering, the steering power consumption is hardly negligible in the overall energy consumption, and therefore, in addition to the longitudinal characteristics of the vehicle, it is necessary to take into account the lateral dynamics of the vehicle in a coordinated optimization process, which all pose higher challenges to the prior art.
Through the above analysis, the problems and defects of the prior art are as follows: the prior art scheme mainly focuses on civil wheeled vehicles, generally decouples the speed and energy collaborative optimization problem into two dimensions of longitudinal speed planning and steering tracking, and cannot be applied to tracked vehicles adopting speed difference steering. However, in the existing tracked vehicle-oriented tracking technology, the energy optimization problem during track tracking is not usually considered, which causes unnecessary energy loss of the vehicle in the path following process, cannot realize the cooperative optimization of the speed and the energy of the tracked vehicle, and seriously affects the fuel economy of the vehicle. .
The difficulty in solving the above problems and defects is: the dynamic and energy flow characteristics of the transmission system of the tracked vehicle need to be comprehensively considered, the related parameters are more, the coupling of each parameter is strong, the degree of freedom of the constructed optimization problem is more, and the solving difficulty is higher.
The significance of solving the problems and the defects is as follows: the hybrid tracked vehicle can fully consider the energy loss during tracking on the premise of ensuring the path following with certain precision, and the optimal balance between high-precision tracking and high-energy-saving tracking is achieved.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method for cooperatively optimizing the speed and the energy of a hybrid tracked vehicle.
The invention is realized in such a way that a hybrid tracked vehicle speed and energy collaborative optimization method comprises the following steps:
step one, designing a coordinate transformation method to convert the GPS coordinates of the original set track into the coordinates of a vehicle local coordinate system so as to facilitate the application of a subsequent algorithm; and converting the data collected by the GPS into data convenient for algorithm application.
Constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle; and constructing a high-precision model containing the dynamics and energy flow characteristics of the hybrid tracked vehicle, and providing a controlled object for the design of a subsequent control algorithm.
And step three, a reinforcement learning method is utilized, tracking errors and energy consumption are fully considered, so that possible behaviors of slipping, side shifting, oversteering and the like in the path following process of the vehicle are reduced, and speed and energy collaborative optimization in the path following process of the hybrid tracked vehicle is realized. The problem of speed and energy collaborative optimization in the hybrid tracked vehicle path following process is solved.
Further, in the first step, the GPS coordinates of the original predetermined track are converted into coordinates of a local coordinate system of the vehicle, and the specific process is as follows:
the GPS longitude and latitude-two-dimensional plane coordinate system-coordinate transformation of the vehicle local coordinate system, firstly, the GPS coordinate of the path is transformed by a Gauss-Kruger projection method, the GPS longitude and latitude coordinate information of the path is processed to obtain the corresponding two-dimensional plane coordinate information, and the coordinate transformation from the GPS longitude and latitude to the two-dimensional plane coordinate system is completed.
Further, the coordinate transformation of the GPS longitude and latitude-two-dimensional plane coordinate system-vehicle local coordinate system specifically includes:
let XOY be a two-dimensional planar coordinate system, X*O*Y*Is a local coordinate system, and is characterized in that,
Figure BDA0003399840600000031
the coordinate of the origin of the local coordinate system is under a two-dimensional plane coordinate system; the coordinate of the point A in the two-dimensional plane coordinate system is (x)A,yA) After conversion to the local coordinate system, the coordinate is (x)A *,yA *) (ii) a The rotation angle of the local coordinate system relative to the two-dimensional plane coordinate system is phi, and the rotation in the counterclockwise direction is positive, the plane coordinate transformation of a can be expressed as follows:
Figure BDA0003399840600000032
in the path tracking problem, the origin of the local coordinate system is fixed to the unmanned crawlerCentroid, and head direction as local coordinate system X*Axial positive direction, then in equation (1)
Figure BDA0003399840600000033
And
Figure BDA0003399840600000034
and (x ', y') is the error of the vehicle in the local coordinate system, and the error in the two-dimensional plane coordinate system is converted into the vehicle local coordinate system, so that the trace tracking control of the vehicle is facilitated.
Further, in the second step, constructing a hybrid tracked vehicle speed and energy collaborative optimization model, including: vehicle kinematics modeling, dynamics modeling, and energy management models.
Further, the vehicle kinematics modeling specifically comprises:
the modeling of the kinematics of a tracked vehicle is relatively complex, with longitudinal dynamics in which the resistance includes the rolling resistance FgUphill resistance FpAnd air resistance FwThe three can be respectively calculated to obtain and calculate the total resistance FrThe following are:
Figure BDA0003399840600000041
in which theta' is the slope, CdIs the air resistance coefficient, A is the windward area;
the dynamic modeling of the tracked vehicle considers the influence of the slip, the slip and the slip of the track, and firstly, mathematical models are respectively established in the straight driving direction and the rotating direction based on the whole vehicle dynamic equation; because the calculation of the details of the ground surface of the crawler belt can lead the calculation amount of the model to be increased sharply, the slip-slip effect of the whole contact surface is concentrated to the centers of the two crawler belts;
straight-line driving direction:
Figure BDA0003399840600000042
in the formula, m and IzRepresenting the reconditioning mass and heading moment of inertia, v and
Figure BDA0003399840600000043
respectively representing the speed and acceleration of the position of the centroid of the tracked vehicle;
Figure BDA0003399840600000044
is the coefficient of ground adhesion, FNA support force for the ground to the tracked vehicle, here equal to the vehicle weight; ftMaximum adhesion to the ground, when Fdrive≤FtThe vehicle does not slip; f1And F2Respectively representing wheel-side driving forces, T, provided by two side driving motors1And T2Respectively representing output torques, i, of motors on both sides0Representing the wheel-side transmission ratio, r being the radius of the driving wheel, etaTIs the efficiency of the motor shaft to the track,
Figure BDA0003399840600000045
namely the longitudinal required power when the vehicle runs;
the rotating direction is as follows:
Figure BDA0003399840600000046
wherein f is1And f2Respectively representing the rolling resistance of the left and right crawler belts, and f representing the rolling resistance coefficients of the crawler belts on the two sides; b represents the center distance of the tracks on both sides, MvehIs the steering resistance moment of the tracked vehicle, g is the gravity acceleration, L is the ground connection length of the track,
Figure BDA0003399840600000051
steering power required for turning the vehicle;
Figure BDA0003399840600000052
for the steering resistance coefficient, the value of which is related to the type of ground on which the tracked vehicle is travelling and the steering radius, the following can be usedThe empirical formula is calculated to obtain:
Figure BDA0003399840600000053
in the formula (I), the compound is shown in the specification,
Figure BDA0003399840600000054
the maximum steering resistance coefficient of the tracked vehicle is represented and determined by the running road surface characteristics of the tracked vehicle, and the maximum steering resistance coefficient of different road surface types can be obtained by looking up a table; r represents the steering radius of the crawler, and the calculation formula of the steering radius is as follows:
Figure BDA0003399840600000055
in the formula, viAnd voRespectively representing the speeds of the inner and outer tracks during steering;
in summary, the total power demand P for vehicle operationdemEqual to power required for longitudinal travel
Figure BDA0003399840600000056
And power required for steering
Figure BDA0003399840600000057
And the sum is provided by a motor on two sides.
Further, the tracked vehicle tracking process kinematic modeling:
according to the motion process of the crawler, the motion state equation can be expressed as follows:
Figure BDA0003399840600000058
in the formula, x, y and theta respectively represent the coordinate of the centroid of the crawler in a geodetic coordinate system and a rotation angle relative to the geodetic coordinate system;
the centroid linear velocity and the rotation angular velocity of the vehicle can be obtained according to the crawler velocities on the two sides:
Figure BDA0003399840600000059
the vehicle is located at the origin of the global coordinate system at the initial moment, is located at the (x, y) of the global coordinate system after the time t, and is represented by P (x, y, theta); let the position of the target to be tracked be represented as P under the global coordinate systemtarget(xtarget,ytarget,θtarget) Using equation (1), the path tracking error of the unmanned tracked vehicle can be expressed as:
Figure BDA0003399840600000061
substituting the formula (7) into a plane coordinate conversion formula (1), and converting the path tracking error into a vehicle local coordinate system:
Figure BDA0003399840600000062
the related path tracking problem is essentially an optimization problem, so that the point position closest to the current position of the vehicle on the reference path is set as a tracking target point to calculate a tracking error, thereby reducing the calculation amount; due to XvehThe direction is the heading of the vehicle in the vehicle coordinate system, so the path tracking error is defined as the sum of the lateral error and the angular error, i.e.:
Figure BDA0003399840600000063
further, the energy management model is constructed, specifically:
the series hybrid moving tracked vehicle model is used, and the power required by the vehicle running is provided by an engine-generator set and output power together;
the output voltage and electromagnetic torque relationship of the generator in the EGS can be expressed as:
Figure BDA0003399840600000064
wherein, UgAnd IgIs the output voltage and current of the generator, wgAnd TgIs the generator speed and torque; keAnd KxIs the electromotive force coefficient and the resistivity; according to the torque balance, the dynamic relation between the engine and the generator in the EGS satisfies the following conditions:
Figure BDA0003399840600000065
wherein, JeAnd JgRotational inertia of the engine and the generator, respectively; in addition, the engine fuel consumption is expressed as a look-up function of engine torque and speed,
Figure BDA0003399840600000071
the battery module is represented by a first-order internal resistance model as follows
Figure BDA0003399840600000072
Wherein R isbatAnd VocRespectively, the internal resistance and open circuit voltage of the battery; i isbatIs the battery current, positive when discharging, negative when charging; qbatIs the battery capacitance, ηbatIs the charge-discharge efficiency of the battery, PbatIs the battery output power; sign (I)bat) Is a sign function whose value can be expressed as
Figure BDA0003399840600000073
Because the EGS, the battery and the driving motor controller are all electrically connected with the bus, the voltages of the EGS, the battery and the driving motor controller are equal; thus, the power distribution of the powertrain satisfies the following relationship:
Figure BDA0003399840600000074
wherein, PdemIs the power demand, P, for the vehicle to travelgThat is, the output power of the generator, which is known from the introduction of the dynamic model, includes the straight-going power and the steering power.
Further, in the third step, a depth certainty strategy gradient algorithm is applied to construct and solve a collaborative optimization problem, and the collaborative optimization problem is divided into two stages, namely offline learning and online application;
an off-line learning stage: according to the speed planning model of the hybrid tracked vehicle defined in the step two, the two-dimensional horizontal abscissa x of the vehicle from the preview pointerrorLongitudinal coordinate y of two-dimensional value of vehicle distance from the preview pointerrorCorner of
Figure BDA0003399840600000075
The running speed v and the turning speed w of the vehicle are used for representing the running state of the vehicle in track following and the rotating speed n of the engineeAnd the battery SOC is used for representing the state of the hybrid power system, and the state is selected as a DDPG state variable, namely
Figure BDA0003399840600000076
Due to the state stThe value ranges of the seven parameters are different greatly, which is not beneficial to network training, so that the selected state parameters need to be normalized and converted into the value range [0,1 ] in a unified way]Or [ -1,1 [)](ii) a Similarly, the torque T of the motor at the inner side of the steering is selectediAnd outer motor torque ToEngine torque TeAs a DDPG algorithm action variable, i.e. at＝{Ti,To,Te}; in addition, the DDPG algorithm feedback reward contains 4 main items: tracking error, energy loss, slip constraint and side shift constraint to ensure optimal balance of fuel economy and vehicle stability in the vehicle track following process and finally realize cooperative optimization of vehicle speed and energy in the track following process; thus the feedback reward setting in the DDPG algorithm is
Figure BDA0003399840600000081
Where κ and λ are maxima, corresponding to v0 and F in trainingdriveThe restriction is carried out, the former prevents unnecessary energy loss caused by vehicle slip, and the latter ensures that the vehicle does not laterally move in the steering process and ensures the driving stability of the vehicle; inputting the selected state variables, action variables and feedback rewards into a DDPG algorithm embedded with a hybrid tracked vehicle speed planning model, and training a strategy network and an evaluation network in the DDPG until the algorithm converges;
because the related system is complex, in order to improve the algorithm training and executing efficiency, a parallel type architecture is adopted to train the algorithm; the CPU cores available in the workstation are first allocated in the appropriate proportion to MATLAB parallel workers, which are of two types: a simulation worker and a learning worker; each learning working device has a copy of the proxy parameters, including the DDPG algorithm and the parameters of the internal network thereof, and each working device not only has a copy of the proxy parameters but also comprises the constructed speed planning model; at each iteration of the algorithm, the worker runs the simulation model and will calculate the state s obtainedtAnd action atPrize rtAnd the network weight parameter is stored in the data buffer area, and meanwhile, each learner synchronously retrieves and extracts the related data of the previous iteration in the data buffer area so as to calculate and update the network weight coefficient; finally, the network parameters calculated by the learner are transmitted into the simulator to carry out the next iterative calculation, and the iteration is carried out circularly until the preset iterative times are reached;
and (3) an online application stage: global path information, initial state variable
Figure BDA0003399840600000082
Obtaining the optimal action variable, namely the optimal control variable, as the input of the trained DDPG algorithm, finally inputting the optimal control variable sequence into the constructed speed planning model of the hybrid tracked vehicle for state iterative update, thereby obtaining the optimal speed sequences at the two sides of the tracked vehicle, and because the vehicle behaviors of the tracked vehicle, such as straight movement, steering and the like, are controlled by the speed difference of the left and right tracked vehicles, the optimal speed sequences are obtained by the control of the speed difference of the left and right tracked vehiclesThe running speed sequences of the tracks on the two sides of the tracked vehicle are optimized, so that the behaviors of the vehicle such as slip, slip and oversteer in the path following process are improved, and the speed and energy collaborative optimization of the tracked vehicle is realized.
It is another object of the present invention to provide a program storage medium receiving user inputs, the stored computer program causing an electronic device to execute the hybrid tracked vehicle speed and energy co-optimization method comprising the steps of:
step one, designing a coordinate transformation method to convert the GPS coordinates of the original set track into the coordinates of a vehicle local coordinate system so as to facilitate the application of a subsequent algorithm;
constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle;
and step three, a reinforcement learning method is utilized, tracking errors and energy consumption are fully considered, so that possible behaviors of slipping, side shifting, oversteering and the like in the path following process of the vehicle are reduced, and speed and energy collaborative optimization in the path following process of the hybrid tracked vehicle is realized.
Another object of the present invention is to provide a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of the hybrid-tracked-vehicle speed and energy co-optimization method.
By combining all the technical schemes, the invention has the advantages and positive effects that: the speed difference steering characteristic of the tracked vehicle is combined, the transverse and longitudinal dynamics and the energy flow characteristic of the vehicle are considered, and the behaviors of slipping, sliding, over-steering and the like of the vehicle in the path following process are improved by optimizing the running speed sequences of the tracks on the two sides of the tracked vehicle, so that the speed and the energy of the tracked vehicle are cooperatively optimized.
Drawings
FIG. 1 is a flow chart of a hybrid tracked vehicle speed and energy co-optimization method provided by an embodiment of the present invention.
Fig. 2 is a schematic diagram of planar coordinate transformation provided by the embodiment of the present invention.
FIG. 3 is a schematic representation of the dynamics of a tracked vehicle provided by an embodiment of the present invention.
Fig. 4 is a schematic view of path tracking of an unmanned crawler according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Aiming at the problems in the prior art, the invention provides a method, medium and equipment for cooperatively optimizing the speed and the energy of a hybrid tracked vehicle, and the invention is described in detail by combining the attached drawings.
Those skilled in the art can also implement the hybrid tracked vehicle speed and energy collaborative optimization method provided by the present invention by adopting other steps, and the hybrid tracked vehicle speed and energy collaborative optimization method provided by the present invention of fig. 1 is only one specific example.
As shown in fig. 1, a hybrid tracked vehicle speed and energy collaborative optimization method provided by the embodiment of the invention includes:
s101: a coordinate transformation method is designed to convert the GPS coordinates of the original set track into the coordinates of a vehicle local coordinate system so as to facilitate the application of the subsequent algorithm.
S102: and constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle.
S103: by utilizing the reinforcement learning method, tracking errors and energy consumption are fully considered, so that the possible behaviors of slipping, side shifting, oversteering and the like in the path following process of the vehicle are reduced, and the speed and energy collaborative optimization in the path following process of the hybrid tracked vehicle is realized.
In S101 provided by the embodiment of the present invention, the GPS coordinates of the original predetermined trajectory are converted into coordinates of a local coordinate system of the vehicle, and the specific process is as follows:
the GPS longitude and latitude-two-dimensional plane coordinate system-coordinate transformation of the vehicle local coordinate system, firstly, the GPS coordinate of the path is transformed by a Gauss-Kruger projection method, the GPS longitude and latitude coordinate information of the path is processed to obtain the corresponding two-dimensional plane coordinate information, and the coordinate transformation from the GPS longitude and latitude to the two-dimensional plane coordinate system is completed.
On the basis, the coordinate change scheme shown in FIG. 2 is adopted to realize the transformation from the two-dimensional plane coordinate system to the vehicle local coordinate system with the centroid of the tracked vehicle as the origin.
The specific scheme is shown in figure 2: XOY is a two-dimensional planar coordinate system, X*O*Y*Is a local coordinate system, and is characterized in that,
Figure BDA0003399840600000111
is the coordinate of the origin of the local coordinate system in the two-dimensional plane coordinate system. The coordinate of the point A in the two-dimensional plane coordinate system is (x)A,yA) After conversion to the local coordinate system, the coordinate is (x)A *,yA *). The rotation angle of the local coordinate system relative to the two-dimensional plane coordinate system is phi, and the rotation in the counterclockwise direction is positive, the plane coordinate transformation of a can be expressed as follows:
Figure BDA0003399840600000112
in the path tracking problem, the origin of a local coordinate system is fixed at the centroid of the unmanned crawler, and the direction of the vehicle head is a local coordinate system X*Axial positive direction, then in equation (1)
Figure BDA0003399840600000113
And
Figure BDA0003399840600000114
and (x ', y') is the error of the vehicle local coordinate system, and the error of the two-dimensional plane coordinate system is converted into the vehicle local coordinate system, so that the trace tracking control of the vehicle is more convenient.
In S102 provided by the embodiment of the present invention, constructing a hybrid tracked vehicle speed and energy collaborative optimization model includes: vehicle kinematics modeling, dynamics modeling, and energy management models.
The two-dimensional planar vehicle dynamics of the simplified tracked vehicle is illustrated in fig. 3. The modeling of the kinematic dynamics of the tracked vehicle is relatively complex, and the resistance in the longitudinal dynamics mainly comprises rolling resistance FgUphill resistance FpAnd air resistance FwThe three can be respectively calculated to obtain and calculate the total resistance FrThe following are:
Figure BDA0003399840600000115
in which theta' is the slope, CdIs the air resistance coefficient, and A is the frontal area.
The dynamic modeling of the tracked vehicle considers the influence of the slip, the slip and the slip of the track, and firstly, a mathematical model is respectively established in the straight driving direction and the rotating direction based on the whole vehicle dynamic equation. The calculation amount of the model is increased sharply due to the details of calculating the track contact surface, and therefore the slip and slip effect of the entire contact surface is concentrated on the centers of the two tracks.
Straight-line driving direction:
Figure BDA0003399840600000121
in the formula, m and IzRepresenting the reconditioning mass and heading moment of inertia, v and
Figure BDA0003399840600000122
representing the speed and acceleration of the crawler centroid position, respectively.
Figure BDA0003399840600000123
Is the coefficient of ground adhesion, FNThe support force given to the tracked vehicle for the ground is here equal to the vehicle weight. FtMaximum adhesion to the ground, when Fdrive≤FtWhile the vehicle is notSlip occurs. F1And F2Respectively representing wheel-side driving forces, T, provided by two side driving motors1And T2Respectively representing output torques, i, of motors on both sides0Representing the wheel-side transmission ratio, r being the radius of the driving wheel, etaTIs the efficiency of the motor shaft to the track,
Figure BDA0003399840600000124
namely the longitudinal required power when the vehicle runs.
The rotating direction is as follows:
Figure BDA0003399840600000125
wherein f is1And f2The rolling resistances of the left and right side caterpillars are respectively represented, and f represents the rolling resistance coefficients of the both side caterpillars. And B represents the center distance of the two side tracks. MvehIs the steering resistance moment of the tracked vehicle, g is the gravity acceleration, L is the ground connection length of the track,
Figure BDA0003399840600000126
the steering power required for turning the vehicle.
Figure BDA0003399840600000127
The steering resistance coefficient, the value of which is related to the type of the ground on which the tracked vehicle runs and the steering radius, can be calculated by the following empirical formula:
Figure BDA0003399840600000128
in the formula (I), the compound is shown in the specification,
Figure BDA0003399840600000129
the maximum steering resistance coefficient of the tracked vehicle is represented and determined by the running road surface characteristics of the tracked vehicle, and the maximum steering resistance coefficient of different road surface types can be obtained by looking up a table. R represents the steering radius of the tracked vehicle, and according to fig. 3, the steering radius calculation formula is as follows:
Figure BDA00033998406000001210
in the formula, viAnd voRespectively representing the speeds of the inner and outer tracks when turning.
In summary, the total power demand P for vehicle operationdemEqual to power required for longitudinal travel
Figure BDA0003399840600000131
And power required for steering
Figure BDA0003399840600000132
And the sum is provided by a motor on two sides.
Modeling the kinematics of the tracking process of the tracked vehicle:
according to the motion process of the crawler, the motion state equation can be expressed as follows:
Figure BDA0003399840600000133
in the formula, x, y and theta respectively represent the coordinates of the centroid of the crawler in the geodetic coordinate system and the rotation angle relative to the geodetic coordinate system.
The centroid linear velocity and the rotation angular velocity of the vehicle can be obtained according to the crawler velocities on the two sides:
Figure BDA0003399840600000134
fig. 4 is a schematic diagram of path tracking of an unmanned tracked vehicle, where the vehicle is located at the origin of the global coordinate system at the initial time, and is located at the global coordinate system (x, y) after a time t, and is characterized by P (x, y, θ). Let the position of the target to be tracked be represented as P under the global coordinate systemtarget(xtarget,ytarget,θtarget) Using equation (1), the path tracking error of the unmanned tracked vehicle can be expressed as:
Figure BDA0003399840600000135
substituting the formula (7) into a plane coordinate conversion formula (1), and converting the path tracking error into a vehicle local coordinate system:
Figure BDA0003399840600000136
the involved path tracking problem is essentially an optimization problem, and therefore the point on the reference path closest to the current position of the vehicle is set as a tracking target point to calculate a tracking error, thereby reducing the amount of calculation. Due to XvehThe direction is the advancing direction of the vehicle in the vehicle coordinate system, so the path tracking error in the scheme is defined as the sum of the lateral error and the angle error, namely:
Figure BDA0003399840600000141
energy management model construction
Using the series hybrid tracked vehicle model, the power required for vehicle travel is provided by both the engine-generator set (EGS) and the output power.
The output voltage and electromagnetic torque relationship of the generator in the EGS can be expressed as:
Figure BDA0003399840600000142
wherein, UgAnd IgIs the output voltage and current of the generator, wgAnd TgIs the generator speed and torque. KeAnd KxAre the electromotive force coefficient and the resistance coefficient. According to the torque balance, the dynamic relation between the engine and the generator in the EGS satisfies the following conditions:
Figure BDA0003399840600000143
wherein, JeAnd JgThe rotational inertia of the engine and the generator, respectively. In addition, the engine fuel consumption is expressed as a look-up function of engine torque and speed,
Figure BDA0003399840600000144
the battery module is represented by a first-order internal resistance model as follows
Figure BDA0003399840600000145
Wherein R isbatAnd VocRespectively, the internal resistance of the battery and the open circuit voltage. I isbatIs the battery current, which is positive when discharged and negative when charged. QbatIs the battery capacitance, ηbatIs the charge-discharge efficiency of the battery, PbatIs the battery output power. sign (I)bat) Is a sign function whose value can be expressed as
Figure BDA0003399840600000151
Since the EGS, the battery and the drive motor controller are all electrically connected to the bus, the voltages of the three are equal. Thus, the power distribution of the powertrain satisfies the following relationship:
Figure BDA0003399840600000152
wherein, PdemIs the power demand, P, for the vehicle to travelgThat is, the output power of the generator, which is known from the introduction of the dynamic model, includes the straight-going power and the steering power.
In S103 provided by the embodiment of the present invention, a collaborative optimization problem is constructed and solved by applying a Deep Deterministic Policy Gradient (DDPG) algorithm, and the construction is divided into two stages, that is, offline learning and online application.
An off-line learning stage: according to the stepsThe speed planning model of the hybrid tracked vehicle defined in the step 2 is a two-dimensional horizontal abscissa x of the vehicle from a pre-aiming pointerrorLongitudinal coordinate y of two-dimensional value of vehicle distance from the preview pointerrorCorner of
Figure BDA0003399840600000153
The running speed v and the turning speed w of the vehicle are used for representing the running state of the vehicle in track following and the rotating speed n of the engineeAnd the battery SOC is used for representing the state of the hybrid power system, and the state is selected as a DDPG state variable, namely
Figure BDA0003399840600000154
Due to the state stThe value ranges of the seven parameters are different greatly, which is not beneficial to network training, so that the selected state parameters need to be normalized and converted into the value range [0,1 ] in a unified way]Or [ -1,1 [)]. Similarly, the torque T of the motor at the inner side of the steering is selectediAnd outer motor torque ToEngine torque TeAs a DDPG algorithm action variable, i.e. at＝{Ti,To,Te}. In addition, the DDPG algorithm feedback reward contains 4 main items: tracking error, energy loss, slip constraint and side shift constraint to ensure optimal balance of fuel economy and vehicle stability in the vehicle track following process and finally realize cooperative optimization of vehicle speed and energy in the track following process. Thus the feedback reward setting in the DDPG algorithm is
Figure BDA0003399840600000155
Where κ and λ are maxima, corresponding to v0 and F in trainingdriveAnd the restraint is carried out, the former prevents unnecessary energy loss caused by vehicle slip, and the latter ensures that the vehicle does not laterally move in the steering process and ensures the driving stability of the vehicle. And then, inputting the selected state variables, action variables and feedback rewards into a DDPG algorithm embedded with a hybrid tracked vehicle speed planning model, and training a strategy network (operator network) and an evaluation network (cognitive network) in the DDPG until the algorithm converges.
Due to the complexity of the system involved, the calculation is improvedThe method adopts a parallel type structure to train the method and the execution efficiency. The CPU cores available in the workstation are first allocated in the appropriate proportion to MATLAB parallel workers, which are of two types: simulation worker and study worker. Each learning worker has a copy of the proxy parameters, including the parameters of the DDPG algorithm and its internal network, and each worker has not only a copy of the proxy parameters but also contains the constructed velocity planning model. At each iteration of the algorithm, the worker runs the simulation model and will calculate the state s obtainedtAnd action atPrize rtAnd storing the network weight parameters into a data buffer, and simultaneously, synchronously retrieving and extracting the related data of the previous iteration by each learner from the data buffer so as to calculate and update the network weight coefficient. And finally, transmitting the network parameters obtained by the calculation of the learner into the simulator, performing the next iterative calculation, and repeating the iterative calculation until the preset iterative times are reached.
And (3) an online application stage: global path information (two-dimensional local coordinates converted by GPS signals), initial state variables
Figure BDA0003399840600000161
The optimal control variable sequence is input into a constructed hybrid tracked vehicle speed planning model for state iterative updating so as to obtain the optimal speed sequences at two sides of the tracked vehicle.
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and its modules of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of hardware circuits and software, e.g., firmware.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. A hybrid tracked vehicle speed and energy collaborative optimization method, characterized in that the hybrid tracked vehicle speed and energy collaborative optimization method comprises:
converting a GPS coordinate of an original set track into a coordinate of a vehicle local coordinate system by adopting a coordinate conversion method;
constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle;
and step three, realizing the cooperative optimization of the speed and the energy in the path following process of the hybrid tracked vehicle by using a reinforcement learning method.
2. The method for collaborative optimization of speed and energy of a hybrid tracked vehicle according to claim 1, wherein in the first step, the specific process of converting the GPS coordinates of the original predetermined trajectory into the coordinates of the local coordinate system of the vehicle comprises: the GPS longitude and latitude-two-dimensional plane coordinate system-coordinate transformation of the vehicle local coordinate system, the GPS coordinate of the path is transformed by a Gauss-Kruger projection method, the GPS longitude and latitude coordinate information of the path is processed to obtain the corresponding two-dimensional plane coordinate information, and the coordinate transformation from the GPS longitude and latitude to the two-dimensional plane coordinate system is completed.
3. The hybrid tracked vehicle speed and energy collaborative optimization method according to claim 2, wherein the coordinate transformation of the GPS longitude and latitude-two-dimensional plane coordinate system-vehicle local coordinate system is specifically: XOY is a two-dimensional planar coordinate system, X*O*Y*Is a local coordinate system, and is characterized in that,
Figure FDA0003399840590000011
the coordinate of the origin of the local coordinate system is under a two-dimensional plane coordinate system; the coordinate of the point A in the two-dimensional plane coordinate system is (x)A,yA) After conversion to the local coordinate system, the coordinate is (x)A *,yA *) (ii) a The rotation angle of the local coordinate system relative to the two-dimensional plane coordinate system is phi, and the rotation in the counterclockwise direction is positive, then the plane coordinate conversion of A is expressed as follows:
Figure FDA0003399840590000012
in the path tracking problem, the origin of a local coordinate system is fixed at the centroid of the unmanned crawler, and the direction of the vehicle head is a local coordinate system X*The axis is forward, then
Figure FDA0003399840590000013
And
Figure FDA0003399840590000014
and (x ', y') representing the error of the actual position and the target position of the vehicle in the global value system, wherein the error is in a local coordinate system of the vehicle.
4. The hybrid tracked vehicle speed and energy collaborative optimization method according to claim 1, wherein in the second step, constructing a hybrid tracked vehicle speed and energy collaborative optimization model comprises: vehicle kinematics modeling, dynamics modeling, and energy management models.
5. Hybrid tracked vehicle speed and energy co-optimization method according to claim 4, characterized in that the vehicle kinematic modeling is in particular: the modeling of the kinematics of a tracked vehicle is relatively complex, with longitudinal dynamics in which the resistance includes the rolling resistance FgUphill resistance FpAnd air resistance FwRespectively calculating to obtain and calculate the total resistance FrThe following are:
Figure FDA0003399840590000021
in which theta' is the slope, CdIs the air resistance coefficient, A is the windward area;
straight-line driving direction:
Figure FDA0003399840590000022
in the formula, m and IzRespectively representing the preparation mass and the course rotary inertia of the tracked vehicle, and respectively representing the speed and the acceleration of the centroid position of the tracked vehicle;
Figure FDA0003399840590000023
is the coefficient of ground adhesion, FNA support force for the ground to the tracked vehicle, here equal to the vehicle weight; ftMaximum adhesion to the ground, when Fdrive≤FtThe vehicle does not slip; f1And F2Respectively representing wheel-side driving forces, T, provided by two side driving motors1And T2Respectively representing output torques, i, of motors on both sides0Indicating wheelSide transmission ratio, r is the radius of the driving wheel, etaTIs the efficiency of the motor shaft to the track,
Figure FDA0003399840590000024
the longitudinal required power when the vehicle runs is obtained;
the rotating direction is as follows:
Figure FDA0003399840590000025
wherein f is1And f2Respectively representing the rolling resistance of the left and right crawler belts, and f representing the rolling resistance coefficients of the crawler belts on the two sides; b represents the center distance of the tracks on both sides, MvehIs the steering resistance moment of the tracked vehicle, g is the gravity acceleration, L is the ground connection length of the track,
Figure FDA0003399840590000031
steering power required for turning the vehicle;
Figure FDA0003399840590000032
the steering resistance coefficient is calculated by an empirical formula:
Figure FDA0003399840590000033
in the formula (I), the compound is shown in the specification,
Figure FDA0003399840590000034
the maximum steering resistance coefficient of the crawler is represented, R represents the steering radius of the crawler, and a steering radius calculation formula is as follows:
Figure FDA0003399840590000035
in the formula, viAnd voRespectively representing the speeds of the inner and outer tracks during steering;
total power demand P for vehicle operationdemEqual to power required for longitudinal travel
Figure FDA0003399840590000036
And power required for steering
Figure FDA0003399840590000037
And the sum is provided by a motor on two sides.
6. A hybrid tracked vehicle speed and energy co-optimization method according to claim 4, characterized in that the tracked vehicle tracking process kinematic modeling: according to the motion process of the crawler, the motion state equation is expressed as follows:
Figure FDA0003399840590000038
in the formula, x, y and theta respectively represent the coordinate of the centroid of the crawler in a geodetic coordinate system and a rotation angle relative to the geodetic coordinate system;
obtaining the centroid linear velocity and the rotation angular velocity of the vehicle according to the crawler velocities on the two sides:
Figure FDA0003399840590000039
the vehicle is located at the origin of the global coordinate system at the initial moment, is located at the (x, y) of the global coordinate system after the time t, and is represented by P (x, y, theta); the position of the target to be tracked is represented as P in the global coordinate systemtarget(xtarget,ytarget,θtarget) And the path tracking error of the unmanned crawler is expressed as:
Figure FDA0003399840590000041
and converting the path tracking error into a vehicle local coordinate system:
Figure FDA0003399840590000042
the path tracking error is defined as the sum of the lateral error and the angular error:
Figure FDA0003399840590000043
7. the hybrid tracked vehicle speed and energy collaborative optimization method according to claim 4, wherein the energy management model is constructed by: the series hybrid moving tracked vehicle model is used, and the power required by the vehicle running is provided by an engine-generator set and output power together;
the output voltage and electromagnetic torque relationship of the generator in the EGS is expressed as:
Figure FDA0003399840590000044
wherein, UgAnd IgIs the output voltage and current of the generator, wgAnd TgIs the generator speed and torque; keAnd KxIs the electromotive force coefficient and the resistivity; according to the torque balance, the dynamic relation between the engine and the generator in the EGS satisfies the following conditions:
Figure FDA0003399840590000045
wherein, JeAnd JgRotational inertia of the engine and the generator, respectively; in addition, engine fuel consumption is expressed as a look-up function of engine torque and speed:
Figure FDA0003399840590000046
the battery module is represented by a first-order internal resistance model:
Figure FDA0003399840590000051
wherein R isbatAnd VocRespectively, the internal resistance and open circuit voltage of the battery; i isbatIs the battery current, positive when discharging, negative when charging; qbatIs the battery capacitance, ηbatIs the charge-discharge efficiency of the battery, PbatIs the battery output power; sign (I)bat) Is a sign function, and is expressed as
Figure FDA0003399840590000052
Because the EGS, the battery and the drive motor controller are all electrically connected with the bus, the power distribution of the power assembly meets the following relation:
Figure FDA0003399840590000053
wherein, PdemIs the power demand, P, for the vehicle to travelgOutputting power for the generator.
8. The hybrid tracked vehicle speed and energy collaborative optimization method according to claim 1, wherein in the third step, a depth certainty strategy gradient algorithm is applied to construct and solve the collaborative optimization problem, and the collaborative optimization problem is divided into two stages, namely offline learning and online application;
an off-line learning stage: according to a defined speed planning model of the hybrid tracked vehicle, a two-dimensional horizontal abscissa x of the vehicle from a pre-aiming pointerrorLongitudinal coordinate y of two-dimensional value of vehicle distance from the preview pointerrorCorner of
Figure FDA0003399840590000054
The speed v at which the vehicle is travelling,the turning speed w of the vehicle is used for representing the running state of the vehicle in track following and the rotating speed n of the engineeAnd the battery SOC is used for representing the state of the hybrid power system, selecting the state as a DDPG state variable,
Figure FDA0003399840590000055
normalizing the selected state parameters, and uniformly converting the state parameters into a value range [0,1 ]]Or [ -1,1 [)](ii) a Similarly, the torque T of the motor at the inner side of the steering is selectediAnd outer motor torque ToEngine torque TeAs a DDPG algorithm action variable, at＝{Ti,To,Te}; feedback reward setting in DDPG algorithm
Figure FDA0003399840590000056
Wherein κ and λ are maxima;
training by adopting a parallel architecture; the available CPU cores in the workstation are first allocated to MATLAB parallel workers in appropriate proportions, and the workers include two types: a simulation worker and a learning worker; each learning worker has a copy of the agent parameters, including the DDPG algorithm and the parameters of the internal network thereof; at each iteration of the algorithm, the worker runs the simulation model and will calculate the state s obtainedtAnd action atPrize rtAnd storing the network weight parameter into a data buffer area; each learner synchronously retrieves and extracts related data of the previous iteration in the data cache region, and calculates and updates the network weight coefficient; finally, the network parameters calculated by the learner are transmitted into the simulator to carry out the next iterative calculation, and the iteration is carried out circularly until the preset iterative times are reached;
and (3) an online application stage: global path information, initial state variable
Figure FDA0003399840590000061
As the input of the trained DDPG algorithm, obtaining the optimal action variable and the optimal control variable, and finally inputting the optimal control variable sequence into the constructed speed planning model of the hybrid tracked vehicle for performing the routineAnd updating the state iteration to obtain the optimal speed sequence of the two sides of the tracked vehicle.
9. A program storage medium for receiving user input, the stored computer program causing an electronic device to perform the method for hybrid tracked vehicle speed and energy co-optimization according to any one of claims 1 to 8, comprising the steps of:
step one, designing a coordinate transformation method to convert the GPS coordinates of the original set track into the coordinates of a vehicle local coordinate system so as to facilitate the application of a subsequent algorithm;
constructing a speed and energy collaborative optimization model of the hybrid tracked vehicle;
and step three, a reinforcement learning method is utilized, tracking errors and energy consumption are fully considered, so that possible behaviors of slipping, side shifting, oversteering and the like in the path following process of the vehicle are reduced, and speed and energy collaborative optimization in the path following process of the hybrid tracked vehicle is realized.
10. A computer device, characterized in that it comprises a memory and a processor, said memory storing a computer program which, when executed by said processor, causes said processor to carry out the steps of the hybrid-tracked-vehicle speed and energy co-optimization method according to any one of claims 1 to 8.
CN202111495143.0A 2021-12-08 2021-12-08 Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment Pending CN114103926A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111495143.0A CN114103926A (en) 2021-12-08 2021-12-08 Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111495143.0A CN114103926A (en) 2021-12-08 2021-12-08 Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment

Publications (1)

Publication Number Publication Date
CN114103926A true CN114103926A (en) 2022-03-01

Family

ID=80364537

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111495143.0A Pending CN114103926A (en) 2021-12-08 2021-12-08 Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment

Country Status (1)

Country Link
CN (1) CN114103926A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114415522A (en) * 2022-03-29 2022-04-29 北京理工大学 Tracked vehicle platform path tracking control method and system
CN115495997A (en) * 2022-10-28 2022-12-20 东南大学 New energy automobile ecological driving method based on heterogeneous multi-agent deep reinforcement learning

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104217615A (en) * 2014-09-16 2014-12-17 武汉理工大学 System and method for preventing pedestrians from collision based on vehicle-road cooperation
CN109017809A (en) * 2018-08-27 2018-12-18 北京理工大学 A kind of energy distributing method based on the prediction of cross-country operating condition
CN109583409A (en) * 2018-12-07 2019-04-05 电子科技大学 A kind of intelligent vehicle localization method and system towards cognitive map
CN109795502A (en) * 2018-09-27 2019-05-24 吉林大学 Intelligent electric automobile path trace model predictive control method
US20200254992A1 (en) * 2019-02-11 2020-08-13 Ford Global Technologies, Llc Lap learning for vehicle energy management optimization
CN112158045A (en) * 2020-09-14 2021-01-01 东南大学 Active suspension control method based on depth certainty strategy gradient
CN112550272A (en) * 2020-12-14 2021-03-26 重庆大学 Intelligent hybrid electric vehicle hierarchical control method based on visual perception and deep reinforcement learning
CN112776817A (en) * 2020-12-30 2021-05-11 三一汽车制造有限公司 Crawler vehicle, control method, controller, and computer-readable storage medium

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104217615A (en) * 2014-09-16 2014-12-17 武汉理工大学 System and method for preventing pedestrians from collision based on vehicle-road cooperation
CN109017809A (en) * 2018-08-27 2018-12-18 北京理工大学 A kind of energy distributing method based on the prediction of cross-country operating condition
CN109795502A (en) * 2018-09-27 2019-05-24 吉林大学 Intelligent electric automobile path trace model predictive control method
CN109583409A (en) * 2018-12-07 2019-04-05 电子科技大学 A kind of intelligent vehicle localization method and system towards cognitive map
US20200254992A1 (en) * 2019-02-11 2020-08-13 Ford Global Technologies, Llc Lap learning for vehicle energy management optimization
CN112158045A (en) * 2020-09-14 2021-01-01 东南大学 Active suspension control method based on depth certainty strategy gradient
CN112550272A (en) * 2020-12-14 2021-03-26 重庆大学 Intelligent hybrid electric vehicle hierarchical control method based on visual perception and deep reinforcement learning
CN112776817A (en) * 2020-12-30 2021-05-11 三一汽车制造有限公司 Crawler vehicle, control method, controller, and computer-readable storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
孔泽慧: "基于强化学习与工况预测的能量管理策略研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, no. 07 *
邹渊;陈锐;侯仕杰;胡晓松;: "基于随机动态规划的混合动力履带车辆能量管理策略", 机械工程学报, no. 14 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114415522A (en) * 2022-03-29 2022-04-29 北京理工大学 Tracked vehicle platform path tracking control method and system
CN115495997A (en) * 2022-10-28 2022-12-20 东南大学 New energy automobile ecological driving method based on heterogeneous multi-agent deep reinforcement learning
CN115495997B (en) * 2022-10-28 2024-01-30 东南大学 New energy automobile ecological driving method based on heterogeneous multi-agent deep reinforcement learning

Similar Documents

Publication Publication Date Title
Xu et al. Torque optimization control for electric vehicles with four in-wheel motors equipped with regenerative braking system
Yang et al. A stochastic predictive energy management strategy for plug-in hybrid electric vehicles based on fast rolling optimization
CN103863318B (en) A kind of hybrid vehicle energy-conservation forecast Control Algorithm based on car-following model
CN104881030B (en) Unmanned vehicle side Longitudinal data tracking and controlling method based on fast terminal sliding formwork principle
CN114103926A (en) Hybrid tracked vehicle speed and energy collaborative optimization method, medium and equipment
CN104340218A (en) Real-time fuel consumption estimation
CN104554251A (en) Hybrid car energy-saving prediction control method based on road gradient information
CN114312845A (en) Deep reinforcement learning type hybrid electric vehicle control method based on map data
CN113911103B (en) Hybrid power tracked vehicle speed and energy collaborative optimization method and system
Goodarzi et al. Stability enhancement and fuel economy of the 4-wheel-drive hybrid electric vehicles by optimal tyre force distribution
Yang et al. Energy management of the power-split hybrid electric city bus based on the stochastic model predictive control
Daoud et al. Path-following and adjustable driving behavior of autonomous vehicles using dual-objective nonlinear MPC
Brembeck Model Based Energy Management and State Estimation for the Robotic Electric Vehicle ROboMObil
Guo et al. Deep reinforcement learning-based hierarchical energy control strategy of a platoon of connected hybrid electric vehicles through cloud platform
Zhang et al. A Novel Model Predictive Control Based Co-Optimization Strategy for Velocity Planning and Energy Management of Intelligent PHEVs
Zhang et al. A survey of vehicle dynamics modeling methods for autonomous racing: Theoretical models, physical/virtual platforms, and perspectives
Masouleh et al. Fuel minimization for a vehicle equipped with a flywheel and battery on a three-dimensional track
Gao et al. Multi-physical cooperative control of plug-in hybrid electric vehicles via cyber hierarchy and interactional network
CN110723207B (en) Intelligent automobile model prediction steering controller based on model reconstruction and control method thereof
Liu et al. Reinforcement learning-based predictive control for autonomous electrified vehicles
Guo et al. Co-optimization of speed planning and power management for automated HEVs in unstructured road scenarios
Zhu et al. Real-time co-optimization of vehicle route and speed using generic algorithm for improved fuel economy
Zhang et al. Trajectory tracking control of driverless racing car under extreme conditions
Yu et al. Model predictive control of a power-split hybrid electric vehicle system with slope preview
Lee et al. Development of an optimal power-distribution-management algorithm for four-wheel-drive electric vehicles

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination