DE102022127339A1 - Method for planning the movement of a truck containing a tank - Google Patents

Abstract

The invention relates to a method for planning the movement of a truck (12) that has a tank (10), the behavior of a fluid (16) in the tank (10) being determined by means of a model of fluid dynamics, in order to determine at least one (10) determining the acting force (18) which is induced by the moving fluid (16), and planning the movement as a function of the at least one determined force (18).

Description

The invention relates to a method for planning the movement of a truck comprising a tank according to patent claim 1.

Trucks with a tank, in particular in a tanker, which are used for the transport of fluids, in particular liquids, are prone to tipping over due to sloshing of the fluid in the tank. Depending on driving maneuvers, a force of the fluid moving in the tank can have a negative effect on driving.

the EP 2 236 863 A2 shows a system for a manual transmission of a tanker truck.

The object of the present invention is to provide a method for planning the movement of a truck that includes a tank, in which a force generated by fluid moving in the tank is taken into account in a particularly advantageous manner, in order to reduce the probability, for example, of a rollover when the truck is driving .

This object is achieved according to the invention by the subject matter of the independent patent claim. Advantageous refinements and developments of the invention are specified in the dependent patent claims as well as in the description and in the drawing.

The method according to the invention serves to plan the movements of a truck that includes a tank. The tank of the truck is arranged in particular in a tank truck, which is coupled as a trailer to a tractor, which together form the truck. In the method according to the invention, a fluid dynamics model is used to determine the behavior of a fluid in the tank, in particular in real time, in order to determine from this at least one force acting on the tank, in particular a sloshing force or sloshing forces, which is caused by the Tank-moving or sloshing fluid is induced, and the movement is planned as a function of the at least one determined force.

In other words, an internal dynamic of the medium in the tank, ie the fluid in the tank, is modeled and observed in real time, for example via sensors. Information obtained from this is made available to a movement planning unit, in particular in a virtual driving environment, in order to plan a path or a trajectory which ensures stability of vehicle dynamics.

The invention is based on the finding that without suitable information about the fluid, for example its liquid state, its distribution and/or movement within the tank boundaries, particularly in real time, the problem of ensuring lateral stability and/or longitudinal stability when the truck is driving is difficult. For example, most liquid tanks for chemicals and/or food have a cylindrical shape and a relatively high center of gravity.

The liquid or fluid inside the tank can, in turn, slide up the side of the tank, causing the center of gravity to shift upwards and outwards. Due to the cleaning requirements for tankers, for example, a change in the shape of the tank, for example, is not practicable and is also associated with high costs. Furthermore, a roll stability control and/or a fully electronic stability control (ESC for short) is required for almost all newly approved liquid tanks. However, such an ESC unit used for tanks typically has a conservative design that makes it unattractive for autonomous driving applications. This is mainly due to the fact that a virtual driver or a self-driving motor vehicle does not know, for example, about the dynamics of the liquid being transported. In this case, the virtual driver can be a unit or an algorithm which is carried out, for example, by an electronic computing device which creates a reference trajectory.

In particular, tank trucks that are only partially filled are susceptible to rollover instabilities due to liquid sloshing. Due to the danger that can arise when a tanker truck rolls over, detailed investigations should be carried out into the parameters that affect the stability of partially filled tanker trucks.

In the case of the present invention, the method, the movement planning is made dependent on the dynamic state of the fluid or the liquid in order to solve the problems mentioned.

One advantage of the method is that the at least one force, in particular sloshing force, is determined or used for movement planning or path planning, for example due to a simplification of an interaction model between fluid or liquid and solid body(s) within the tank boundaries, in particular without significant computational effort. The one for the procedure computational effort to be run is important in view of hardware limitations that exist in standard automotive controllers. During motion planning or trajectory planning, the at least one predicted force can be taken into account in order to design an optimal path or trajectory that ensures the stability of the tanker-truck system and, for example, at the same time enables smooth and non-conservative reference values. It also enables the design of tracks that are fuel efficient and/or avoid rollovers, rattling and/or similar phenomena.

In an advantageous embodiment of the invention, an algorithm is used for the movement planning, which takes into account the at least one force in transverse commands and/or longitudinal commands for defining a trajectory in order to reach a threshold value for driving stability of the truck.

In a further advantageous embodiment of the invention, a mechanical substitute model, in particular a pendulum model, is used for a partially filled elliptical tank to determine or calculate the at least one force, in which a pendulum mass describes a movement of the fluid.

In a further advantageous embodiment of the invention, at least one parameter of the substitute model is set or used on the basis of experimentally determined data.

In a further advantageous embodiment of the invention, corresponding characteristic eigenvalues of selected oscillation modes of the substitute model are considered or taken into account for the movement planning.

Further advantages, features and details of the invention result from the following description of a preferred exemplary embodiment and from the drawing(s). The features and combinations of features mentioned above in the description and the features and combinations of features mentioned below in the description of the figures and/or shown alone in the figures can be used not only in the combination specified in each case, but also in other combinations or on their own, without going beyond the scope of the leave invention.

• 1 schematisches Ablaufdiagramm eines Verfahrens zur Bewegungsplanung eines einen Tank umfassenden Lastkraftwagens; und
• 2 schematische Darstellung des Tanks des Lastkraftwagens mit Illustrierung eines Ersatzmodells für das Verfahren gemäß 1.

It shows:

• 1 schematic flowchart of a method for planning the movement of a truck comprising a tank; and
• 2 Schematic representation of the tank of the truck with an illustration of a substitute model for the method according to FIG 1 .

1 shows a schematic flowchart for a method for planning the movement of a truck 12 comprising a tank 10 , the tank 10 being in particular a tanker 14 of the truck 12 . In the method presented, the behavior of a fluid 16 in the tank 10 is determined, in particular in real time, by means of a fluid dynamics model, in order to determine at least one force 18 acting on the tank 10, which is caused by the moving or sloshing fluid 16 is induced, and the movement is planned as a function of the at least one force 18 determined.

For example, parameters 30 of the fluid 16 and/or of the tank 10 are included in the model of the fluid dynamics. Parameters 30 of truck 12 and/or its surroundings can thus be included, for example elasticity of the suspension, the shape of tank 10, and the position of the road.

Finally, the at least one force 18 is determined in a movement planning unit 20 (Motion Planning Unit, MPU for short), the movement planning unit 20 being part of an electrical computing device or being executed on the latter, for example. The electrical computing device is, for example, a control unit of truck 12.

The at least one determined force 18 is now processed, for example, by a trajectory planner 22 of movement planning unit 20, which is used to plan the movement, which is represented by box 24, with an algorithm being used in particular for planning the movement, in particular of trajectory planner 22, which uses the at least a force 18 is taken into account in transverse commands and/or longitudinal commands for setting a trajectory 26 in order to reach a threshold value for driving stability of the truck 12. The threshold simply indicates that, for example, the truck should be prevented from tipping over.

Advantageously, to determine the at least one force 18, a mechanical substitute model, which is used in particular in 2 is outlined and is designed, for example, as a pendulum model, used for a partially filled ellipsoidal tank 10, in which a pendulum mass 28 describes a movement of the fluid 16.

The parameters of the substitute model are determined experimentally, in particular, on the basis of this ter data is set or specified. Advantageously, correspondingly characteristic eigenvalues selected oscillation modes of the substitute model are taken into account for the movement planning.

One of the main applications for the method is the use for driver assistance systems, for example active lane keeping up to level 5 trucks 12 driving fully autonomously.

The stability or dynamics and their prediction for the truck 12 is a problem known from the literature, with the greatest challenge in ensuring the stability of the truck 12 occurs when it performs a dynamic maneuver that can lead to sloshing of the fluid 16. The sloshing forces that act internally as disturbances to the truck 12 depend on several factors, such as the viscosity of the medium, the shape of the tank 10, a shape and/or elasticity of baffles if they are present, the height and the distribution of the fluid 16 in the tank 10.

The latter depends not only on the height of the fluid 16 in the tank 10, but also on how the baffles or other baffles distribute the fluid 16 in the tank 10 in the longitudinal direction. Due to the high variability of the above factors, the at least one force 18 can be very dynamic and extremely difficult to predict.

Without suitable information about the fluid 16, for example its liquid state, its distribution and/or movement within the tank boundaries in real time, ensuring lateral stability and/or longitudinal stability becomes a particularly great challenge. For example, most chemical and/or food liquid tanks are cylindrical in shape and have a relatively high center of gravity. The liquid inside the tank 10 can in turn slide up the side of the tank 10, shifting the center of gravity up and out.

Another challenge is that autonomous vehicles should follow a continuous set of vehicle longitudinal dynamics commands to follow either a specified path and/or trajectory 26, hereinafter referred to as Tracking Reference Medium. A path is a geometric entity that is time-independent and can be a cascade of points and/or a smooth curve with spatial reference coordinates. The trajectory 26, on the other hand, is a time-dependent structure that specifies where the vehicle, ie, for example, the truck 12, should be at a specific point in time.

The path and/or trajectory planning in driver assistance systems and autonomous driving systems is highly dependent on the vehicle model used. In the case of truck-trailer combinations, such as the present truck 12 with the associated tanker 14, the problem of trajectory or movement planning becomes particularly difficult due to the dynamic sloshing behavior of the moving fluid or medium in the tank 10.

Without information about the sloshing of the fluid 16, only extremely conservative paths and/or trajectories 26 can be predicted to avoid instabilities due to braking and/or steering. This limits the performance in predicting or staying in lane, so that there are restrictions on intentional lane departures and/or similar maneuvers.

The method presented here only makes it possible to provide the movement planning unit 20 with information about the dynamics of the liquid charge in particular in real time in order to enable the design of path planning and/or trajectory planning that is safe and/or robust against instabilities.

In the prior art, turbulence and/or sloshing dynamics of the liquid in a moving vehicle are mostly neglected for active and/or passive driving control. In addition, closed solutions for modeling the slosh phenomenon are very computationally intensive.

Instead, the proposed method aims to properly characterize the complicated interaction between the e.g. structural vibrations and the internal tank liquid based on simplified modeling approaches, such as pendulum analogies, and then to provide this information to the motion planning unit 20 to design a suitable trajectory or to create a trajectory 26 without having to make a conservative estimate or reaching the edge of the stability of the truck 12 during its journey.

2 12 shows the tank 10 of the tanker 14 of the truck 12. The tank 10 is partially filled and the fluid 16 is allowed to move freely within the confines of the tank 10. FIG. Furthermore, for the sake of simplicity, it is assumed in the exemplary embodiment shown that within of the tank 10 there are no static, semi-active and/or active baffles counteracting the free movement of the fluid 16 and/or the distribution of its surface.

In order to obtain information about the interaction between the fluid 16 and forces 18 acting on the tank through the slosh phenomenon, a simplified modeling approach is used. In the exemplary embodiment, an approach is selected for modeling the model and thus the at least one acting force 18 in which the liquid or the fluid 16 can slosh freely in the tank 10 .

For example, an equivalent mechanical model for a partially filled, in particular ellipsoidal, tank 10 is used here in order to calculate the at least one force 18 . The pendulum mass 28 simulates the sloshing effect in particular for the first mode of the volume fraction of the liquid or the fluid 16 involved in the sloshing. The solid mass, the pendulum mass 28 represents an inertia and a weight of the remaining fraction of the fluid 16. For the sake of simplicity, the equivalent dynamic Equation of motion not explicitly stated. Parameters of the simplified pendulum model can be adjusted based on experimental investigations. The sloshing force, the at least one force 18, can be calculated as a second-order ordinary differential equation, which depends on a mass and a density of the fluid 16, the radius of the tank 10, the natural frequency and a damping ratio of the first mode form of the liquid vibration.

It should be noted that higher eigenvalues of the liquid sloshing are well above excitation frequencies that arise from the steering of the truck 12, so only the fundamental mode shape is considered.

Depending on a current driving maneuver, an equation solver is proposed to solve the decisive motion equation, which is particularly relevant for lateral stability and the avoidance of rollovers. The solution of this numerical solver can also be predicted in a future time horizon, for example by forward simulation of the model. Consequently, lateral force, pitch and/or roll effects can be calculated in real time and for a future time horizon. This can be referred to as slosh prediction. With this, the trajectory 26 can now be advantageously determined.

In this case, for example, the trajectory 26 can be followed by a driver assistance system, and this can be achieved by actuating the steering and/or the engine and/or the brakes of the truck 12 .

The box 24 the 1 shows, represented by the discrete points, the information about the reference position, the speed and/or the acceleration, the values being calculated in real time taking into account the active force 18 . Here, d _x stands for a forecast horizon of the planned orbit and the predicted sloshing force. The forecast horizon should be chosen in such a way that the propagation of modeling errors is as small as possible. For example, the length of the prediction horizon can be a speed dependent lookup table.

The method shown has the advantage that the force 18 can be predicted for the movement planning unit 20 without any appreciable computing effort, since, for example, the interaction model between liquid and solid has been particularly simplified. The computational effort of the model is particularly important because of the hardware limitations imposed by standard controllers. The at least one force 18 is taken into account when planning the path. This enables the design of a trajectory 26 that is fuel efficient and avoids rollovers and/or chatter and/or similar phenomena. The method can thus particularly advantageously avoid jackknife, a pendulum movement of the tanker 14, a deviation from the lane and/or a rollover during various road maneuvers.

reference list

10

tank

12

truck

14

tank truck

16

Fluid

18

power

20

movement planning unit

22

trajectory planner

24

Crate

26

trajectory

28

pendulum mass

30

parameter

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• EP 2236863 A2 [0003]

Claims (5)

Method for planning the movement of a truck (12) comprising a tank (10), in which a behavior of a fluid (16) in the tank (10) is determined by means of a fluid dynamics model, in order to derive at least one fluid (16) acting on the tank (10). Determine force (18) which is induced by the moving fluid (16), and the movement planning is carried out as a function of the at least one determined force (18). procedure after claim 1 , characterized in that an algorithm is used for the movement planning, which takes into account the at least one force (18) in transverse commands and/or longitudinal commands for defining a trajectory (26) in order to achieve a threshold value for driving stability of the truck (12). procedure after claim 1 or 2 , characterized in that to determine the at least one force (18) a mechanical substitute model for a partially filled ellipsoidal tank (10) is used, in which a pendulum mass (18) describes a movement of the fluid. Device/method according to claim 3 , characterized in that at least one parameter of the replacement model is set using experimentally determined data. Device/method according to claim 3 or 4 , characterized in that vibration modes of the substitute model selected according to characteristic eigenvalues are taken into account for the movement planning.

Images