Model-based force estimation device for a motorized vehicle

Abstract

The invention describes a force estimation device for a motorized vehicle that generates an assistance torque dependent on an external force. The external force is generated directly or indirectly by a user and applied to the vehicle by pushing or pulling. To estimate this external force, a model-based state estimation is used. Based on the model parameters and system inputs (tilt angle, motor current), this first predicts the system states for the next discrete point in time. In a second step, this prediction is corrected based on the speed and / or acceleration measurements. This method for model-based force estimation enables a less noisy and more precise estimation than conventional methods that use an inverse transfer function.

Description

Technical FieldThe present invention relates to a power assist device intended to assist a person in pushing or pulling a vehicle by means of an electric motor. Moreover, the invention relates to an electrically assisted vehicle including the aforementioned auxiliary device. The electrified vehicle is further characterized as being actuated by a pulling or pushing force exerted by a person. Examples of such vehicles are motorized (children's) cars, (bicycle) trailers, scooters, push trucks, skateboards or bag trucks.Prior ArtPower assist vehicles in which the driver of the vehicle is assisted by an electric motor are becoming increasingly important. This trend is clearly visible, in particular in the area of personal mobility. A known example of power-assist vehicles are electric assist bicycles (e-bikes) that measure and boost torque applied to the crank by an electric motor. The development of E-bikes has so far been receiving much attention. However, the force-supporting concept can also be applied to other, less conventional vehicles. In the following, therefore, force-supporting vehicles which are set in motion by a pulling or pushing force will be considered. These include, in particular, (children's) cars, (bicycle) trailers, scooters, scotchers, skateboards, push trucks or bag trucks, in which an electric motor assists the force applied to the vehicle by the user or the user in order to simplify the conveyance of people or cargo. In order to be able to electrically support the force introduced by the human being onto the vehicle, however, this must first be quantified.In the case of E-bikes, the drive torque is conventionally measured with the aid of torque sensors in the pedal crank (WO 2015 055 673 A1). More recent inventions address the problem of torque quantification in E-bikes with the aid of model-based methods (US 2006 095 191 A1, EP 3 782 895 B1, US 10 625 818 B2). In this case, a dynamic model of the bicycle is first derived and the total torque which is the cause of the measured speed and speed change is then calculated with the aid of speed and road inclination measurements. Taking into account the measured motor torque on the basis of the measured motor current, a torque can then be determined which can be associated with the torque applied by the human being. This general, model-based approach is also referred to below as a virtual torque sensor.In the external force estimation of other motorized vehicles, such as trailers, there are also approaches to determine the driving force. Analogously to the torque sensor in the bottom bracket of the bicycle, force sensors in the drawbar (WO 2021 150 160 A1) or in the handle can be used in a towed or pushed vehicle. Other approaches use elastic elements in the suspension or the drawbar in order to be able to draw conclusions about the external force via the measurement of the deformation of the elastic element (WO 2017 144 832 A1). Finally, decoupling between a towing vehicle and a trailer is also possible in order to enable the trailer to follow without transmitting force to the towing vehicle (WO 2017 140 626 A1).The consideration of virtual force sensors, i.e. a model-based estimate of the external force in vehicles which are being towed or pushed, is described in U.S. Pat. No. 10,625,818 B2, WO 2018 189 621 A1, JP 6 324 335 B2 and DE 10 2017 217 650 A1. The method described in US10625818B2 is based on a transfer function which describes the dynamic behavior of the vehicle. The transfer function takes into account the inertia of the vehicle and the load, the stick and slip friction, the drag and the gravitational forces resulting from the slope of the road. By forming the inverse transfer function, the total force with which the vehicle was driven and, taking into account the motor current, also the proportion of the external force is calculated. For the force measurement, only one speed sensor (e.g. by means of Hall sensor systems in the wheel) and one acceleration sensor (for measuring the road inclination) are required. Furthermore, the patent U.S. Pat. No. 10,625,818 B2 describes a parameter identification based on a sequential least squares algorithm. The parameter identification is intended to make it possible to take account of the time-variant behavior of the system, for example in the case of a changing luggage load, in the transfer function.The principle for determining the external force set forth in WO 2018 189 621 A1 likewise uses a dynamic model of the vehicle in combination with a measurement unit, but ignores explicit consideration of the air resistance.The approach described in JP 6 324 335 B2 additionally dispenses with the use of a speed sensor and instead uses a speed observer for estimating the motor rotational speed on the basis of the motor voltage and the motor current. Then, the estimated angular velocity is given to a disturbance observer, which estimates the total external force (disturbance). The disturbance observer consists of an inverse function of the dynamic behavior of the vehicle, similar to the inverse transfer function in US 10 625 818 B2.After the external force estimation, admittance control is implemented, which provides a reference speed for a PID speed controller, which adjusts the motor voltage so as to achieve the desired target speed. The admittance control is to keep the required load on the operating person of the vehicle as constant as possible. For this purpose, the load for moving the vehicle is defined in the form of a virtual mass and a virtual damping factor.The method described in DE 10 2017 217 650 A1, similar to the aforementioned patents, determines the external force using a mass value, a friction value, the measured motor torque and an acceleration measured by inertial sensors.In contrast to the use of virtual force sensors, the methods described in DE 10 2021 206 057 A1 and DE 10 2018 209 496 A1 determine force components, by means of which a drive current is calculated for compensating the slope-driven force and / or rolling friction. DE 10 2021 206 057 A1 additionally presents a method for estimating the vehicle mass and the loading mass. The mass estimation is carried out either by using a least-square estimator or a Kalman filter or by periodic excitation of the motor and subsequent use of the Goertzl algorithm. DE 10 2018 209 496 A1 describes the estimation of the mass of the vehicle by means of a recursive algorithm based on the least squares of errors during the braking process. In addition, in this disclosure, a detection unit is provided that detects the tilt angle of a baby carriage by using trigonometry and measured acceleration of the transport device.Disadvantages Including DisadvantagesThe implementation of conventional physical force and torque sensors increases the complexity of the system, resulting in higher costs. The sensors require a more expensive hardware setup, which makes both the production process and the installation more expensive. The necessary periodic maintenance and possible repairs can result in additional costs.Another disadvantage is the limited adaptability of conventional sensors. Especially in vehicles which are pushed or pulled, force sensors allow only the measurement of the force effect on the vehicle at only one point (e.g. either on the drawbar or the handle). This prevents modifications in the manner of use if, for example, a hanger is no longer to be pulled but rather is to be pushed. In such a case, a second sensor would have to be used in order to be able to measure the force effect at the desired point, which in turn increases the complexity of the system.Model-based force estimates or virtual force sensors, on the other hand, are cost-effective and detect external force actions independently of the point of application (pulling or pushing).However, the virtual force sensors described for force-supporting vehicles (not e-bikes) have some disadvantages.The aforementioned approaches are based on the use of an inverse transfer function. In the implementation given in US 10 625 818 B2, the use of this inverse function results in a noisy force estimate. The reason for this is that within the inverse transfer function, the second derivative of the motor position must be determined on the basis of the Hall sensor system of the motor. Since the Hall sensors only provide a quantized, discrete measurement of the rotor position, the second derivative (even with a high number of poles and gear ratio) and thus also the calculated external force is very noisy. In order to be able to use the estimated force signal for the regulation of the motor, this would have to be filtered by a low-pass filter with a low cut-off frequency, which leads to a marked delay in the support. The approach disclosed in JP 6 324 335 B2 avoids this problem by estimating the speed with an observer. However, this approach is less accurate, since the information of the Hall sensors, which are available in most motors in any case, is not taken into account. The use of acceleration sensors-as described, for example, in WO 2018 189 621 A1 and DE 10 2017 217 650 A1-likewise permits a direct measurement of the acceleration in the direction of travel, wherein the measurement can, however, be impaired by measurement noise, vibrational influences and drift and does not take into account the known prior knowledge about the acceleration from the measurement of the wheel speed.One of the most important aspects in order to be able to correctly estimate the externally introduced force is the reliable determination of the road inclination even under high horizontal accelerations of the vehicle. U.S. Pat. No. 10,625,818 B2 proposes an inclination angle sensor based on an acceleration sensor. However, the sole use of acceleration sensors for angle-of-inclination estimation leads to estimation errors in the presence of accelerations in the direction of travel. EP 3 782 895 B1 therefore proposes a method which weights the measured values of an acceleration sensor and of a rotation rate sensor as a function of the vehicle acceleration determined by Hall sensors. The latter patent is limited only to estimating the driving torque in an E-bike.The object is to provide a method for performing a processFor the regulation of electrically assist vehicles that are pulled or pushed by an external force, a quantification of the external force is required. For economic and practicable reasons, quantification should be carried out without additional force sensors. The challenge is to estimate the external force as quickly and noise-free as possible to enable direct and seamless assistance of this force by the electric motor of the vehicle.SolutionThe object is achieved by a motorized vehicle having the features of the main claim. The vehicle is characterized in that it is pushed or pulled by an external force. In addition, the vehicle has at least one electric drive. The electric drive can be realized as a wheel hub motor or as a motor with a mechanical coupling to a wheel.The electric drive is controlled in such a way that an assistance torque dependent on the external force is applied. In this case, the user can independently select the degree of assistance by actuating the operating unit.According to the invention, the electric drive has a calculation unit for estimating the external force, wherein a calculation model in the form of the driving dynamics of the vehicle is stored in the calculation unit. The calculation unit also takes into account the pitch inclination of the vehicle in addition to the speed and the acceleration of the vehicle and the motor torque generated by the electric motor in its calculation model. By taking into account the pitch inclination of the vehicle, the driving resistance due to a roadway inclination corresponding to the pitch inclination, for example in the event of a hill rise, is taken into account in the calculation model, among other things.Furthermore, the calculation unit is characterized in that it determines the external force with the aid of a state estimate, such as an extended or an "unscented" Kalman filter. The state estimation is based on the model of the driving dynamics of the vehicle and the model parameters and the sensor measured values. Based on the nonlinear model in combination with the model parameters, the measured motor current, the measured speed (and acceleration) and the measured road inclination, the internal states are estimated according to the principle of state estimation, and the force externally introduced onto the vehicle is derived via this. This is possible by considering the externally introduced force in the model not as a system input, but as an additional, internal system state.The road gradient is preferably measured by a 3D acceleration sensor in combination with a rotation rate sensor. The 3D acceleration sensor is capable-at constant speed-of measuring the gravitational acceleration and thus of determining the orientation of the sensor and thus of the vehicle. However, in the case of a (positive or negative) acceleration of the vehicle, the estimate would deviate from the actual pitch angle. In this case, however, the geometric length of the three-dimensional acceleration vector also deviates from the length of the gravitational acceleration vector, which is why in this case the rotation rate sensor can additionally be used to correct the estimation of the pitch angle.Preferably, the calculation unit is configured to estimate the external force based on a total mass of the vehicle, the total mass including the vehicle mass and the payload. The total mass can preferably be determined by means of a weight sensor, by means of input information and / or by means of a predefined constant value. Preferably, a weight sensor is provided for determining the driver's mass, which weight sensor is implemented in the wheel suspension of the vehicle. For a further preferred embodiment for determining the total mass without the use of an additional sensor, a parameter identification is provided during operation. In particular in the case of sharp increases, a changed mass has a high influence on the model behavior. This fact is preferably to be used to adapt the model parameter of the mass such that the quadratic error between model and process is minimized.In a preferred configuration, the calculation unit is configured to estimate the externally applied force from aerodynamic drag data. These drag data include drag resulting from the speed of the vehicle and drag due to any wind. The resistance caused by the speed of the vehicle, in particular the wind resistance, contributes substantially to the overall resistance. The inclusion of these drag data in the calculation model helps minimize discrepancies between the predicted external force and the actual external force.Preferably, it is intended to detect during sliding vehicles whether the latter is oriented parallel to the road or whether it is tilted by external action-as is often the case with baby carriages. This is necessary to determine whether the pitch angle of the vehicle corresponds to the road inclination. This detection can preferably be carried out by sensor systems in the handle or by evaluating the measurement profile of the rotation rate, acceleration and speed sensor systems.The invention thus provides an electrified vehicle in which the pushing or pulling external force exerted by a user or a user can be estimated with high accuracy and thus also supported.AdvantagesThe use of a model-based method for determining the externally applied force offers various advantages. On the one hand, no additional hardware in the form of a force sensor is required. Further, an external force is detected regardless of whether the vehicle is pulled or pushed. The explicit use of state estimation furthermore offers the advantage that different sensor modalities can be easily fused to one another. For example, a direct measurement of the forward acceleration with the aid of an acceleration sensor can be taken into account. However, the force estimation also works without a measurement of the acceleration and only on the basis of the Hall sensor measurement, since the acceleration represents an internal state of the system and can be derived from the speed. Furthermore, the state estimation allows automatic low-pass filtering of the estimated signal depending on the selection of the covariance matrices. By appropriately choosing covariance matrices, a good compromise can be found between estimation speed and noise suppression.Enumeration of the FiguresThe invention is explained in more detail on the basis of an exemplary embodiment which is illustrated in the drawings. It shows: FIG. 1 : Bicycle trailer with an electric drive FIG. 2 : Model of the vehicle and state estimationThe numbering legend is as follows: 10 Vehicle 11 Wheels 12 Motor 13 Drawbar 14 Handle 15 Calculation unit 16 Measuring unit 17 Battery 20 Bicycle 21 Clutch 30 Model 31 Driving force 32 Current 33 Pitch angle 34 Air resistance generated by wind 35 Speed 36 Acceleration 37 Sensor values 40 State estimate 41 Inputs 42 Prediction model 43 System state Speed 44 System state Acceleration 45 Discrete point in time 46 Measurement model 47 Output of the measurement model 48 Second output of the measurement model 49 Vector of all measurement model outputs 50 Comparison of the predicted sensor values with the measured sensor values 51 Correction factor 52 Corrected system states 53 Conversion of acceleration into force 54 Estimated external forceEMBODIMENT OF THE INVENTIONFIG. 1 shows a motorized vehicle ( 10). The vehicle can either be pulled over a drawbar (13) or pushed over a handle (14). The vehicle can also be coupled to another vehicle by a clutch ( 21), so that the pulling force is not necessarily exerted directly by a human, but is transmitted indirectly via the rear wheel of another vehicle, e.g. a bicycle ( 20), to the motorized vehicle.For electromotive assistance, the vehicle has at least one electric drive ( 12), which can drive the wheels ( 11) (or the wheel) of the vehicle. The electric drive can also have a speed-force transmission in the form of a transmission. The motors are supplied with electrical energy by a battery (17).The external force required for driving the electric motor or motors is estimated by a calculation unit ( 15) mounted on the vehicle. For estimating the external force, a measuring unit ( 16) for measuring the pitch angle of the vehicle and thus indirectly for measuring the road inclination is furthermore required. This measurement unit can be, for example, an inertial measurement unit (IMU) that enables 3D acceleration and rotation measurement.A block diagram for estimating the external force is shown in FIG. 2 and will be explained in more detail below.In order for the calculation unit ( 15) to be able to estimate the external force, a model ( 30) is first required, which describes the driving dynamics of the motorized vehicle. One possible model is given by the following equation:Here, the measurable model outputs are described by the speed v(t) ( 35) and the acceleration ( 36) of the vehicle ( 10). Furthermore, m tot characterizes the total mass of the vehicle ( 10) including the payload, J w the rotational inertia of the wheels ( 11), r w the radius of the wheels ( 11), and g the gravitational acceleration. The rolling friction resistance of the wheels (11) is taken into account by μ, whereas the sliding friction and static friction of the bearings of the drive train are taken into account by β 1 and β 0. The air resistance is determined using the air density p, the aerodynamic drag coefficient c d and the front surface A of the vehicle. Furthermore, the model has four inputs: 1. the external driving force F ext(31). 2. the current I M(32) of the motor (12). Via this, the generated driving force can be determined in consideration of the engine constant kt, the gear ratio y, the efficiency coefficient of the engine c M and the tire radius r W. 3. the pitch angle θ (33) of the vehicle and thus the gradient angle of the road. 4. the air resistance is generated by wind F Wind(34), which acts on the vehicle.In order to correctly adjust the motor current (32) and thus the driving force, an estimation of the external force F ext according to the main claim of the invention is required.A state estimate (40), for example an (extended or unscented) Kalman filter, uses in a first step the state information x k-1( the underslope identifies a vector) of a system at the discrete point in time k-1 in combination with a prediction model (42) and its inputs u k(41), in order to give a prediction about the state at the point in time k (45). In a second step, the measured system variables z k are calculated using a measurement model ( 46) on the basis of the predicted system states. These hypothetical measured values are then compared ( 50) with the actual sensor values ( 37) and used via a correction factor ( 51) to correct the predicted states. The result is the estimate of the internal system states ( 52) at the current point in time k, which estimate is corrected by measured values, it is assumed that both the measured values and the state values predicted by the model are subject to a multivariate, average-free Gaussian noise (λ k or κ k) respectively.In order to be able to ascertain the external force F ext by such a state estimation in the present example, the model equation described above will first be transformed into a discrete system of the shape. In addition, the external force F ext(31) previously considered as input is converted to an internal state of the system, since state estimates are unable to estimate system inputs. The latter is realized by defining a further, inner state whose input consists of white noise free from mean values. This additional state represents the acceleration a ext caused by the external force. The discrete system equations extended by a second state are thus obtained as:These two equations are used in the state estimation in order to make a prediction of the two states x 1,k= v k(43) and x 2,k= a k(44) for the time k on the basis of the states at the time k-1 and the inputs of the system (motor current, pitch angle).For the correction of the estimate on the basis of measurement signals, the measurement model ( 46) of the system is required in the second step. The speed v of the vehicle represents the first output of the measurement model ( 47). The output function is described by:If the acceleration of the vehicle can additionally also be measured, for example by an acceleration sensor or the differentiation of the speed, this can be used as a second output of the measurement model ( 48) in order to correct the state estimate in the next step. The corresponding output function is:The correction ( 51) takes place after the comparison of the calculated with the measured sensor values based on the covariance matrices of the noise vectors λ k or κ k. These covariance matrices of the measurement and process noise are adjustment parameters and their choice is relevant for the convergence and dynamics of the state estimation.In summary, the idea is that the state estimate determines acceleration of the vehicle caused by all known sources (motor torque, gravitational force due to inclination, frictional resistance, air resistance, etc.) and estimates the remaining acceleration portion required to give the measured speed and acceleration. This external acceleration component is attributable to the external force and can be directly converted from the estimated state x 2.k= a ext,k into the estimated external force ( 54) as follows ( 53):It should be clear that other design embodiments of the motorized vehicle, other model equations and other types of state estimation are also possible in comparison with the described embodiment without departing from the scope of protection of the main claim.

Claims

Force estimation device for a motorized vehicle (10) which generates an electric assistance torque as a function of an external force, wherein the external force is generated directly or indirectly by a user or a user and acts on the vehicle from the outside, wherein this external force is quantized on a model basis, wherein the force estimation device comprises: a. at least one output for presetting an electric current (32) or a voltage to a motor (12) which can exert an electric torque for moving the vehicle b. an input for transmitting the pitch inclination of the vehicle (16) c. an input for transmitting the motor speed and / or movement speed (35) and / or the vertical acceleration of the vehicle (36) d. a calculation unit (15) which estimates the force acting on the vehicle from the outside, wherein the estimation is characterized in that it is carried out on the basis of a dynamic model (30) of the vehicle, the speed and pitch inclination (33) and using a state estimation (40), wherein the state estimation uses a prediction model (42) in a first step and a measurement model (46) in a second step in order first to predict (45) the internal states of the system on the basis of the motor specifications and then to correct (50, 51, 52) this prediction on the basis of the input values.The power estimation device for a motorized vehicle according to any one of the preceding claims, wherein the vehicle is configured to have the externally applied power applied thereto in a pushing or pulling manner.The power estimation apparatus for a motorized vehicle according to any one of the preceding claims, wherein the state estimation is realized by a normal, extended or unscented Kalman filter.Force estimation device for a motorized vehicle according to one of the preceding claims, wherein the pitch inclination of the vehicle is measured and transmitted by a 3D acceleration sensor and / or a 3D rotation rate sensor.The force estimation device for a motorized vehicle according to claim 4, wherein at low horizontal accelerations, the information of the acceleration sensor is weighted more, and at high horizontal accelerations, the information of the rotation rate sensor is weighted more.The power estimation device for a motorized vehicle according to any one of the preceding claims, wherein the vehicle speed is measured and transmitted based on the Hall sensor system of the engine in consideration of the transmission ratio or by Hall sensors mounted directly on the wheel of the vehicle or another vehicle coupled to the vehicle.The power estimation apparatus for a motorized vehicle according to any one of the preceding claims, wherein the vertical acceleration is measured and transmitted either by an acceleration sensor or by differentiation of the speed signal.The power estimation device for a motorized vehicle according to any one of the preceding claims, wherein the calculation unit determines the external power taking into account the total mass (vehicle mass and load mass) of the vehicle, the total mass being defined by input information and / or by a predetermined constant value and / or being measured by a weight sensor mounted in the wheel suspension or on the suspension of the vehicle and / or being determined during the vehicle by means of an online parameter identification that selects the mass so as to minimize the error of the prediction model output (45) and the measured sensor values (37).The power estimation device for a motorized vehicle according to any one of the preceding claims, wherein the calculation unit estimates the external power taking into account the air resistance, the air resistance resulting from the measured speed of the vehicle and / or the estimated wind resistance (34) or measured by a wind sensor.The power estimation device for a motorized vehicle according to any one of the preceding claims, wherein when the powered vehicle is pushed, it is detected whether the vehicle is oriented parallel to the road or whether it is tilted by external action.Force estimation device for a motorized vehicle according to Claim 10, wherein the tilting detection takes place via touch sensors in the handle (14) of the vehicle and / or takes place by evaluating the rotation rate, acceleration and speed sensors.

Images