Untitled record

description

ELECTRIC VEHICLE WITH WHEEL-INDIVIDUAL FRICTION BRAKES AND WHEEL-INDIVIDUAL DRIVE AND BRAKING METHODS FOR IT

The present invention relates to an electric vehicle with a friction brake on each wheel of the vehicle, at least the wheels of an axle of the vehicle are driven individually for each wheel by an electric motor and the electric motor can be used both as a drive and as a regenerative brake, as well as a braking method for such a vehicle.

In electric vehicles, two brakes are usually available, namely a regenerative brake (or drive) and a friction brake, because the regenerative brake can not always swallow the entire kinetic vehicle energy. The combination of the two brakes is to be seen as “a braking process, since the overall effect for the vehicle must be considered. The typical combination includes friction brakes on the wheels, preferably adjustable individually for each wheel, for example an electric or mechatronic brake with the necessary power transmission devices such as gears, levers, rods, ropes, pumps, pressure transmissions (liquids, gases), control devices (electronics, valves) and a regenerative braking drive machine, such as an electric motor in generator mode), a flywheel storage device , a pressure accumulator, etc., for driving certain or all wheels (which are preferably individually controllable). The braking system can be controlled by a control device, for example the ECU (electronic control unit), preferably from hardware and software, in accordance with the requirements, external conditions and / or specifications.

Vehicle stability systems, such as ABS or ESP, are well known from conventional vehicles. Such systems try to keep the vehicle in a stable state by adjusting braking forces for each individual wheel, for example in the event of braking (ABS) or cornering (ESP). Such stability systems are of course also used in electric vehicles. Today, however, the regenerative brake is switched off if necessary and only the friction brake is used. The main reason for this is that, on the one hand, today's ABS and ESP controls are built for the hydraulic brakes and, on the other hand, that the regenerative energy generated during braking has to be absorbed somewhere (usually by the battery). If the battery were suddenly full when braking, it would no longer be able to draw current or the brakes could not be charged for other reasons, such as cold, overheating, etc., which means that braking or braking behavior has an indeterminable effect on the vehicle would suddenly change. Therefore, in today's electric vehicles, for safety reasons, only the friction brake is used in a stability case (ABS, ESP). However, this also means that the friction brake must be designed for the most unfavorable braking situation (e.g. emergency braking, long downhill journeys), although this case only occurs very rarely in real ferry operations.

Thus, in a known electric vehicle, on the one hand, the regenerative energy is not optimally used and, on the other hand, large-sized and disadvantageously heavy friction brakes are necessary, which worsens the use of the available electrical energy (eg in the form of a reduced range).

It is therefore an object of the present invention to eliminate the above-mentioned disadvantages in an electric vehicle by means of a new braking system or braking method.

This object is achieved in that in the event of braking on a driven wheel by the friction brake and / or the electric motor working as a regenerative brake, a basic braking torque can be set, which is superimposed by a modulation torque generated by the electric motor and in the vehicle next to a battery Another energy sink is provided in order to absorb the regenerative energy produced during braking. The energy sink has sufficient absorption potential to absorb the resulting / 18

AT 12 010 U2 2011-09-15 Austrian

Patent Office to be able to absorb regenerative energy. For this purpose, several energy sinks existing next to one another can also be provided in the vehicle, which are used as required. This means that the friction brake only has to generate a certain (small) proportion of the braking torque, which means that the friction brake can be made smaller. If necessary, additional torques (braking or drive torques) are generated by the electric motor. The drive or braking torque of the electric motor (generator) can be regulated very quickly electronically. This advantage is used in the procedure for quick interventions (e.g. in the ABS-ESP case). With this quick adjustment, a locked wheel is made free again or, in the case of a freely rotating wheel, the locking limit is determined, which results in a modulation of the electrical braking torque (or drive torque). Furthermore, all regenerative energy is used in the event of braking, which improves the electrical efficiency of the vehicle.

According to the invention, the wheels of an axle are also driven individually for each wheel when an electric motor drives both wheels of an axle via a differential, since the differential distribution of the electric drive torque (braking torque) to the wheels can be controlled by the differential.

In the event of braking, it can happen that the regenerative energy cannot be absorbed by the battery (as the preferred primary energy sink), for example because it is already fully charged or battery charging is currently not possible for other reasons. A further energy sink must therefore be provided that can absorb the energy that cannot be stored in the battery. The following can be provided as an energy sink:

· Deliberately not fully charging the battery in normal operation in order to be able to safely accommodate regenerative energy in the battery in the event of braking. For example, an additional pre-calculation can be carried out, e.g. after a long uphill drive: The regenerative energy that is later released when driving downhill due to the ongoing braking can be estimated in advance and the battery charge level is kept or brought to a corresponding level. This means that the battery itself is used as a further energy sink, which is deliberately not optimally used in normal operation.

· An electrical heater, also with the generation of tolerable excess temperature (in order to accommodate as much regenerative electrical energy as possible). Switching devices, with which the heat is converted in the heater but kept away from the cabin, can also be used in order to be able to dissipate as much electrical energy as possible.

· An electric air conditioning system, also with the generation of tolerable low temperatures (in order to accommodate as much electrical energy as possible). Switching devices, with which the regenerative energy is converted but the cold is kept from the cabin, can also be used in order to be able to dissipate as much electrical energy as possible.

· The combination of heating and air conditioning to dissipate as much electrical energy as possible for as long as possible.

· Deliberate suboptimal operation of the generator and converter and the associated power electronics up to the limit of the permitted power loss, for example through pulse currents that produce higher losses than average currents. If, for example, instead of an average current, a pulse current with a duty cycle of 1:10 is set, 10 times the pulse current generates 100 times the power loss. Switching frequencies in the converter that are as high as possible can also be used for additional losses or short-term, deliberate deterioration in the circuit can be switched on.

· Combination of generator braking and countercurrent braking: one or several electric travel motors are operated as a generator and provide the electrical

2/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

Energy for countercurrent braking of the other travel motor (s). Countercurrent braking means that (as far as this is permitted by current and torque) the motor is deliberately controlled electrically against its direction of rotation.

· Towing a range-extenders: If an auxiliary motor (internal combustion engine) for battery charging (range-extender) is available, it can be used for engine braking purposes from the electric drive motors (as a generator) via the range-extender generator (as a motor) in towing mode can be used as the engine braking effect of an internal combustion engine.

Of course, any combination of the above energy sinks is also conceivable as a further energy sink.

Normally, the braking force is distributed for good driving behavior or good braking effect, ie a relatively large amount of braking effect at the front, less at the rear. In order to be able to make the brakes smaller (more cost-effective), the distribution of controllable brakes for long, weaker braking (typical downhill driving) can be optimized in favor of the best heat distribution and only redistributed for a favorable braking effect when there is a particular need for braking. In this way, the frictional heat can be distributed more favorably in terms of heat to the friction brakes and thus enables friction brakes with smaller dimensions. With this “favorable distribution of the friction braking torque, comfort can also be taken into account, for example braking torques with noises (eg squeaking) can be avoided by other distribution. Reserves with regard to overheating or wheel slip can also be built up by, for example Certain wheels are deliberately braked less or by deliberately braking less on the front wheels, for example, in order to keep them steerable or more brakable on poor roads (eg ice, snow).

In the event of braking, in particular in the case of heavy braking or in a case of stability, the braking torques of the friction brake and especially the braking torque or drive torque of the regenerative brake must sometimes be adjusted or distributed very quickly. The torque of an electric machine (electric motor, generator) can be adjusted very quickly if the machine current or the magnetic field is adjusted. The electrical machine is therefore predestined for setting such rapid torque changes and thus for modulating the braking torque. The rates of change are limited by inductances, whereby the maximum rate of current change is determined for a given voltage. Compared to mechanical settings such as the mechatronic friction brake with an electric motor, the adjustability and the possible speed of adjustment of the electric machine is much higher and also more comfortable despite the limited current change rate, since no mechanical control noises occur.

However, when the vehicle is in operation, not only do heavy braking and cases of stability occur, but rather “normal braking” is the rule. With such normal braking, slow changes in distribution occur, which is referred to as “biending”. This means that a certain part is braked regeneratively and, depending on the driving speed and braking requirement, a certain braking torque of the friction brake or the regenerative brake is additionally set. The method according to the invention can of course also be used for such a "biending", the following strategies being conceivable:

· The battery can be charged: in this preferred case, the highest possible generator power and the least possible addition of friction brakes are set, taking into account the power limits of the generator, converter and battery. The relationship between regenerative brakes and friction brakes changes, for example, with the driving speed due to the speed-dependent generator effect.

· Battery cannot be charged or cannot absorb all of the electricity produced: as little as possible is supplemented by friction brakes and the energy that cannot be charged into the battery is absorbed by the additional energy sink.

3/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

· Heat-favorable distribution of the friction braking torques.

· Noiseless distribution of the friction braking torques: If certain friction braking torques lead to noises (eg squeaking), the braking torques can be divided differently in order to generate the same total torque with lower-noise friction braking torques. The distribution can be changed within the friction brakes or the distribution of regenerative and friction braking can also be changed.

· Wear-favorable torque distribution: the friction brakes can be distributed according to certain wear criteria, such as uniform wear.

· Overall optimum distribution: instead of the optimum of certain individual criteria, a combined optimum can be set, for example as uniform a heat distribution as possible while largely avoiding noise.

It is also conceivable to switch the regenerative component of a wheel from a braking torque to a drive torque during braking, for example to make a blocked wheel run particularly quickly again in a stability case (faster than would be possible by simply reducing the braking torque).

In conventional brakes, a given brake distribution is first started, which comes from the same hydraulic pressure on all wheels. If the wheels lock or the vehicle leaves the lane, this given distribution changes abruptly to the ABS or ESP state in order to regulate the road guidance forces of individual wheels by locking and letting go of individual wheels. In contrast to this, it is particularly advantageous to provide individual wheel brakes which can each be individually controlled and adjusted. A wheel-specific braking torque can come from a friction brake and / or an electric motor as a drive or brake. In this way, optimal braking torques for each individual wheel can always be generated, whereby a sudden changeover from “specified distribution to ABS-ESP distribution, as with conventional brakes, can be avoided. The optimal individual wheel braking torques can be set under the following criteria:

· Maximum utilization of the road grip on each wheel for the best deceleration.

· Yaw control: A desired yaw rate is determined from the steering wheel position and, by introducing individual wheel forces, a torque is introduced around a yaw axis that triggers the desired yaw rate.

[0030] Within certain limits, a request for deceleration can be fulfilled at the same time as a desire for yaw rate by means of individual wheel moments.

In certain cases, situations can arise in which braking forces arise on certain wheels and drive forces on other. An optimal moment for each individual wheel can also originate from different, sometimes contradicting demands. In this case, a decision algorithm can make a meaningful choice. This can, for example, be the sensible preference for the driver's last action. So if the driver, for example first brakes and then steers, if both wishes cannot be met, preference can be given to the steering request with the existing road grip. In principle, desired limits can also be preferred. If, for example, the driver's wish would cause a rollover to the side (roll) or forwards (nod, for example on a motorcycle), this wish can in principle be limited to a harmless intervention. Physically sensible measures can also be preferred, e.g. To reduce the kinetic energy of the vehicle for an accident as far as possible, for which assistance systems of the vehicle, such as a distance assistant, obstacle detection, etc., can also be used.

If the vehicle is in a driving dynamically problem-free state, which statistically accounts for the vast majority of the time in operation, energy, wear or comfort-favorable optimizations can be carried out, for example as described above.

4/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

Wheel-specific interventions can, however, not only take place in the event of braking, but also while the vehicle is in operation. For example, the following applications are conceivable:

· If, for example, a tire has worn unevenly, ie “has flat spots, then an attempt can be made to compensate for the wear again. In addition, the drive torque can be a little higher in good places on the tire over longer distances and a little lower in bad places (of course, all within the limits of comfort and noise) in order to compensate for the wear.

· If the friction brakes develop different (uneven) effects (e.g. braking torques, heat) over the range of friction, this can be detected by wheel measurements and compensated for with the electrical braking drive torque through temporally coordinated superimposition or even overcompensated, for example to overheat Make less braking and thus achieve a more even distribution again.

Such deliberate small wheel torque fluctuations can also be impressed to improve measurements on the wheel, for example a slip measurement. During normal driving, the wheel torques can be quite small, for example if only the vehicle resistance has to be overcome or if only moments around zero are required due to slight downhill driving. In such cases, small torque fluctuations can improve a wheel slip measurement and thus critical situations (e.g. Snow, ice, gravel), which can also be present on one side, can be recognized more quickly or even before they are noticed by blocked wheels.

The conventional engine braking effect of vehicles with internal combustion engines normally acts on a drive axle via a differential. It is always there and is usually especially used when going downhill. There is little or no additional braking, and the limit of liability is clearly felt through these unequal wheel forces. This behavior can be simulated with a wheel-specific torque control, so that, for example the impression of differential, differential lock or all-wheel drive (even when braking) can be simulated in order to give the driver the familiar feelings for assessing the liability. In this way, for example, a rear-wheel drive can be simulated in an adjustable manner, which oversteers or breaks when accelerating. In general, any type of pre-warning for the driver can be brought about by means of wheel-specific interventions, in particular one that also occurs with conventional drives.

[0038] So far, individual wheel torques have been spoken of. Of course, this results in a wheel force in the direction of travel of the wheel at the wheel support on the roadway over the radius. A force in the running direction combined with a force in the lateral direction (cornering force) is always possible on a vehicle wheel, whereby one or both of them can be even zero. A resultant force arises vectorially from the two normal forces. This can flow into a wheel force-oriented model for controlling the brake. This wheel force-oriented model can lead to “torque vectoring”, i.e. the steering and acceleration or deceleration of the vehicle with force vectors that are introduced at the wheels. This control method works according to the calculation method that the total force vector arises as the vector sum of all effective wheel force vectors. The yaw moment is the sum of all moments around a point, i.e. the sum of the wheel force vectors multiplied by their normal distance to a defined point and the total force vector multiplied by the normal distance to the defined point. The defined point can (but does not have to) be the center of gravity of the vehicle. The yaw moment introduced around the yaw axis through the center of gravity can thus be determined. After this calculation, the individual wheel force vectors can be designed by changing the torque so that the vehicle drives the desired curve (or straight ahead with an infinite curve radius). When limits of detention are exceeded

5/18

AT 12 010 U2 2011-09-15 Austrian

Patent office, the vector calculation must be carried out again with reduced forces in order to maintain the torque vectoring and the stability of the vehicle.

If the wheels can also be steered individually or in groups, the wheel force-oriented model can be expanded to include a wheel-specific steering angle. This means that not only the wheel force in the running direction but also the cornering force can be set on each wheel. If grip limits are exceeded, the vector calculation must be carried out again with the reduced forces and / or other individual wheel steering angles in order to maintain the torque vectoring and the stability of the vehicle. If only a few steering angles can be influenced (for example only front wheel steering), the method naturally works in the same way, with the difference that fewer variables can be influenced. If wheel steering is also intentionally or unintentionally caused by other influences (e.g. deflection and geometry of the wheel suspension), the procedure also works in the same way, with the difference that state-given variables can be taken into account.

Example of advantages of the steering intervention: If, for example, the left side of the vehicle has a good roadway (asphalt) and the right side has a bad roadway (ice), as is quite practical, the “good wheels” could be braked very hard, but those on ice almost not. This introduces a strong yaw moment due to the strong one-sided braking. Without steering intervention, the control unit would have to decide between good deceleration (which would be possible through asphalt) and directional stability, e.g. according to the above method “preference for the last driver's request. Both would not work at the same time. With additional steering intervention, on the other hand, the vehicle can be braked heavily on one side and the direction of travel can be maintained at the same time through slight automatic counter-steering. Both of these are of course only possible within the framework of physical limits.

In order to enable the braking method according to the invention described above, different measurements can also be provided.

If the vehicle has a yaw rate sensor (as is common today with ESP), the yaw rate measurement can be used to determine whether the yaw rate caused by the braking process corresponds to the target value that would be expected from the steering wheel position. Control interventions in the braking process can bring the actual yaw rate closer to the desired one. With this, for example automatic stability can be brought about in the event of a one-sided collision by bringing the vehicle back into a desired state in the event of an unintentional turning by means of the yaw rate control. Another use case is, for example, when towing a trailer (or heavily loaded with unfavorable loads), in which additional forces act on the vehicle. The yaw rate control can try to drive the vehicle in the desired state.

In order to influence the wheel torque very quickly, it must also be possible to measure the wheel speed quickly in a wide speed range. To measure the speed of rotation, toothed disks are currently provided on the wheels, which are usually magnetically scanned. The counting of teeth in a gate time (as is often the case) is only possible for the method according to the invention at high speeds, because then enough teeth pass for sufficient accuracy even in a short gate time (for fast measurement results). At low speeds, such a measurement does not provide a measurement result that is sufficiently accurate for rapid influencing of the wheel torque. For a sufficiently quick and accurate measurement, the time between two teeth can be measured instead. This means that measurements can be made quickly with good resolution even at low speeds, since time measurements are in principle very accurate. The inaccuracy of the tooth spacing has a disadvantage, since it constantly simulates other speeds. Since microprocessors can now be used for measurement, these accuracy errors can, however, be measured and compensated for. In the same way, unequal wheel radii (e.g. due to tolerances of the tire on the rim or more worn spots on the tire) can be filtered out. Spring movements of the wheel can result in a variable wheel speed, whereby this effect can also be filtered out or suppressed if the spring movement is known (actually or through models).

6/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

The wheel slip can be measured comparatively (wheels to each other), changing over time (e.g. if wheel speeds change faster than according to the vehicle model (mass, road forces) or with respect to the vehicle speed over ground (e.g. from the Doppler effect or time behavior of a the lane sent and partially thrown back signal), from other received signals (e.g. Radio station or GPS signal) or from a normal GPS measurement. This wheel slip can provide information on road grip before the driver is surprised by unfavorable grip.

In a conventional hydraulic brake, a uniform contact pressure is brought about by the same pressure on all wheels. In the case of friction brakes that can be set individually for each wheel (eg mechatronic, electrical brakes as described in W02009 / 062880 A1), the set braking torque can be determined using various methods. For this purpose, either the mechanical stress state in the brake calliper can be determined (e.g. by strain sensors), or the braking distance can be determined, which is related to the braking force and thus the braking torque. It is also possible to draw conclusions from the current about the braking force or braking torque via the actuator current. Under certain circumstances, a self-boosting component of the brake can also be taken into account.

In the case of the electrical machine, conclusions can be drawn very effectively about the torque by measuring or modeling the electrical states. When adjustments are made to the electrical machine, conclusions can be drawn about the total torque and thus the contribution of the friction brake from the known contribution of the electrical machine and the reaction (the wheel speed or the wheel slip or other measured wheel values ​​such as deflection of the wheel suspension). In this way, the friction brakes can also be checked or readjusted for their uniform braking effect by determining and comparing the effect (see above) on the wheels with a known (possibly the same on certain or all wheels) electrical braking drive torque.

The determination of the braking torque is also possible by measuring the drag force on the outboard brake lining. With the floating caliper brakes that are frequently used, the outboard brake lining is taken along after it has been pressed onto the brake disc. This entrainment force can be measured as a support force, for example piezoelectrically or by observing the change in length of a spring used for support. If this spring is not very strong (for reasons of cost and space), it can still be determined (by distance measurements or by mere switches or position sensors) how long the spring can support the pad against being carried along and from when the pad against the usual stop wanders. This makes it possible to determine a point on the characteristic between the activation of the friction brake and the friction torque that is passed through when the brake is actuated. From this point, if the brake data is known, conclusions can be drawn about other friction braking torques. For example, in a (usual) simple friction model, the friction force is linearly proportional to the contact pressure and the coefficient of friction. From the observation of the above spring (at a point or in an area) the resulting frictional force can be measured and by means of the friction model it can be concluded which input data lead to a certain frictional force. So, for example from the position (or the motor current of the actuator or other known or accessible data of the friction brake) in a certain state of the above spring (e.g. exactly while the switch is being operated) to the necessary new setting data of the friction brake (e.g. position, motor current, other known or more accessible Data) for newly generated braking torques can be deduced from the friction model (e.g. the simple linear relationship). In this way, predictable braking torque can be achieved despite possible changes (coefficient of friction, parameters in the brake) from an area or point of the spring observation, so to speak, the observable area or point can be extrapolated to larger areas or the entire braking torque range. The models of friction and brake can of course be more precise than just proportional, for example. If both brake pads on the brake disc and both sides of the disc are the same or very similar, it can be assumed that the inboard pad makes a very similar contribution (which can also be determined more precisely from models) and thus the total friction force and at known

AT 12 010 U2 2011-09-15 Austrian

Patent Office ter Geometry the braking torque can be determined. The method can also be seen as a target / actual comparison: The modeling of the brake (with friction lining) yields an expected value for which brake setting the above spring should assume which positions. This is compared with when these spring states actually occur and from the deviation it can be concluded how the model of the brake (with friction lining) must be corrected. With the model corrected in this way, the deviation from the actual conditions is different (meaningfully smaller) and an ever smaller deviation is set in one or more steps by adapting the model.

The braking method according to the invention can also take other circumstances into account. If, for example, there are no electric motors (or other drives with braking and drive properties) on certain wheels, the method in question can be limited to adjustable braking torques on these wheels.

If wheels are driven via an axle differential and there are friction brakes that can be adjusted individually for each wheel, the method in question is nevertheless possible on these wheels. The differential input torque is distributed to two output torques (the gear ratios play a role). The wheel torque is the sum with the correct sign of the torque from the differential and the friction brake (taking into account any gear ratios). With a controllable differential input torque and brakes that can be controlled individually for each wheel, the process-typical wheel-specific braking or drive torque can thus be generated from this context. It is therefore irrelevant for the method from which energy source the drive originates or into which form of energy the drive machine converts the regenerative energy (e.g. electrical, mechanical, pneumatic or hydraulic) in the event of braking.

This means that mixed drives such as internal combustion engines via front axle differential and electric motor (s) on the rear axle can be operated just as typically in the process. In this case, the operating state “driving and braking simultaneously even (or especially) over a certain period of time would be of interest, as it allows energy to be transported from the combustion engine via the road to the electric motors (as a generator) for battery charging, thus avoiding an additional generator on the combustion engine can be.

The wheel suspension can introduce uneven rotational speeds into the wheel when the wheel makes spring movements. It is also possible to choose a suspension that causes particularly strong non-uniform rotation with suspension, e.g. when the wheel is arranged on a swing arm and the drive shaft passes through the pivot point and the wheel axle and drive shaft are connected e.g. via gears. Such irregular rotations can cause effects that can be taken into account in the brake control. If the drive torque remains constant when the spring is moving, the speed changes, but the spring behavior remains unaffected. If the drive speed remains constant when the spring is moved, there are changes in torque. This results in counter forces to the spring behavior and the spring behavior is made more difficult. This can also be set and used so that it corresponds to certain strengths of a shock absorber. If the spring movement has different effects depending on the direction (e.g. less damped compression, more damped rebound), an effect similar to the usual movement direction-dependent shock absorbers is created. With such a simulated shock absorber behavior, an analogy to a conventional shock absorber can be created. The conventional converts the energy withdrawn for damping into heat. In the simulation, the energy can also be drawn electrically and thus used or stored again.

The knowledge that is gained from the braking process via the vehicle can of course also be passed on, for example airbags can be triggered or belts tightened. Road conditions or braking (evasive) vehicles can be communicated to other vehicles (or control stations) (eg via radio). Likewise, for example, the Schein8 / 18

AT 12 010 U2 2011-09-15 Austrian

Patent Office throw setting must be corrected.

In addition to the driver, braking, acceleration or speed specifications can of course also come from other sources, for example from a cruise control. Steering requests can also come from other systems in addition to the driver, for example from an assistance system for avoiding obstacles or lane keeping.

Assistance systems can also be made available with the braking process, such as automatic handbrake and starting aid (Hill Holder) or Hill Descent Control (construction of a “wedge of road material under the wheels when driving downhill in the field).

The method can also automatically respond to special cases such as snow chains (also mounted unevenly, on individual wheels or axles, also with unequal effects such as damaged or different chains) because it notices the new slip and yaw situation or it can Change the settings for this (possibly also switchable by the user). In the sense of the above adaptation of the braking resp. Drive torque on conditions about the wheel circumference can also be responded to by the unequal effect of chains over the wheel circumference (e.g. whether parts of the chain are missing or e.g. the chain is on the road or the tire is right now) by adapting the torque to these conditions .

Other brakes can also be included in the process, such as various retarders (eg trucks).

The method can also ensure stability in vehicles with trailers and semi-trailers against “jack-knife” (buckling). The trailer can be controlled or braked by the process, or only the towing vehicle is additionally stabilized against the trailer forces.

The present invention is explained in more detail below with reference to FIGS. 1 to 7, which schematically show advantageous, non-limiting exemplary embodiments. It shows

1 shows an electric vehicle with a braking system according to the invention,

2 to 6 show the states in a wheel force-oriented model for controlling the brake and

7 shows a preferred measurement of the wheel braking torque.

The greatly simplified and schematically illustrated vehicle 1 in FIG. 1 has four wheels 2_b 2₂, 2₃, 2₄ which are driven individually for each wheel. The front wheels 2_b 2₂ are each driven directly by an electric motor 6 for this purpose. The wheels 2₃, 2₄ the rear axle are driven by a further electric motor 11 via a differential 12. These wheels 2₃,2₄the rear axle could, however, as well as the front wheels 2_b 2₂, be driven directly by their own electric motor. Each wheel 2_b 2₂, 2₃, 2₄ also has a wheel-specific friction brake 4, that is to say an individually controllable and adjustable friction brake 4, which acts on the wheels 2 in a known manner via a brake disk 3 (or brake drum)_b 2₂, 2₃, 2₄ works. A wheel speed sensor 5 (or a sensor for an equivalent measured variable) is also provided on each wheel.

Friction brakes that can be controlled and adjusted individually for each wheel are nowadays preferably electrically driven (also conceivably from other energies), the motor being able to be arranged on the friction brake, but also being able to act remotely. There are usually always connecting elements such as gears, rods, levers, ropes, gases, liquids between the motor and the friction brake. However, all friction brakes can also be operated individually by one or more central motors (electrical or based on other energies) by controlling the distribution, for example via levers or valves. On the steerable wheels 2 ^ 2₂, here the wheels of the front axle, can be known in

9/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

Steer servo motors 13 may also be provided, as indicated in FIG. 1, for example to implement steer-by-wire.

Likewise, a trailer coupling 14 with an interface for signal and energy transmission between a trailer (not shown) and the vehicle 1 can also be provided on the vehicle 1.

An ECU 6 (Electronic Control Unit) controls all functions and components of the brake system and also receives the measurement signals from the various sensors, such as wheel speed sensors 5 or other vehicle sensors 7, such as steering angle, yaw rate, accelerations, speeds, etc., and can also have various control inputs, e.g. for brake pedal position, accelerator pedal position, gear, etc. The ECU 6 could of course also be divided into several units, for example a separate unit for brake control. The power electronics 8 for the individual wheel drives 6, 11 are also controlled by the ECU. The power electronics 8 can of course also be based on several units, for example per axle or per wheel 2₁ 2₂, 2₃, 2₄ be divided.

In the electric vehicle 1, a battery 9 is also provided, which supplies the electric motors 6, 11 with electrical energy via the power electronics 8, or which is provided as a primary energy sink for receiving regenerative electrical energy. In addition, there is a further energy sink 10, as described above, for taking up regenerative energy that cannot be taken up by the battery 9.

In the following, a wheel force-oriented model for controlling the brake is described with reference to FIGS. For easier understanding, a general vehicle 1 in a general driving state is described below. Theoretically, the following explanations are also applicable to any number of wheels in any arrangement, such as single-lane, two-lane, multi-lane, with rigid, steerable (e.g. Articulated steering) or articulated connection (semitrailer, drawbar trailer, trailer with or without steerable wheels). For a simplified explanation, a two-lane vehicle with four wheels 2i, 2₂, 2₃, 2₄ with the main steering of the front wheels (ie a "normal car) shown, in which 2i, 2₂, 2₃, 2₄ a vector force F_vi F_V.2, F_V.3, F_V.4th to the street occurs. This vectorial force F_vi, F_V.2, F_v3, F_V.4th is broken down into a force in the direction of wheel travel F._R1, F_R2, F_R3, F_R4 (Driving or braking force, hereinafter generally referred to as longitudinal wheel force) and a force normal to the direction of wheel travel F_S.i, Fs2, Fs3, Fs4 (cornering force, for example when cornering or driving on a slope, hereinafter generally referred to as lateral wheel force). These wheel vector forces F_V1, F_V2, F_v3, F_V4 are in general neither for each bike 2_b 2₂, 2₃, 2₄the same (not even on both front and rear wheels), nor is the direction of the force vector generally the same. Vehicles usually have 1 front-wheel steering, with the steered wheels, here the two front wheels 2_b 2₂, take favorable and thus different steering angles. In fact, however, all wheels steer 2_n 2₂, 2₃, 2₄ through deflection and forces with small changes in the steering angle. These small steering angle interventions can be zero (e.g. with rigid axles), they can occur due to the geometry of independent wheel suspensions, they can be brought about consciously (e.g. to achieve good cornering behavior) or they can also be set intentionally with adjusting devices (e.g. with electric motors or spring position-dependent mechanics) become. In general, the assumption here is a driver's request for the direction of travel and speed (including acceleration, i.e. possible braking or increase in speed), the general individual steering angle on all wheels 2 ^ 2₂, 2₃, 2₄ and generally different wheel vector forces F_V1, F_V2, F_V3, F_V4 on the wheels 2_b 2₂, 2₃, 2₄ causes.

Imagine, for example, cornering with heavy braking. Due to dynamic wheel load distribution when braking, the front wheels are pressed more strongly against the road than the rear wheels and the centrifugal force presses the outer wheels more strongly than the inner wheels. Therefore every wheel has 2_b 2₂,2₃, 2₄ an individual contact pressure on the road, at the front, the strongest on the outside of the curve and the least on the rear, on the inside of the curve (the rear wheel, the wheel on the inside of the curve, may even lift off). Fig. 3 shows the vehicle 1, which is in a

10/18

AT 12 010 U2 2011-09-15 Austrian

Patent office

Curve brakes, whereby the forces are all shown in one area (wheel contact area) and not in three dimensions. The centrifugal force F is generated by cornering_z and the braking force F by braking_B.. Inertia forces can also be taken into account. These forces act as the resultant force F._R. at a certain point A of the vehicle 1. An equally large total wheel force F_Degree in the opposite direction from all wheels 2_b 2₂, 2₃, 2₄ are applied together to keep the vehicle 1 stable in this state. The centrifugal force F_z and the braking force F_B. arise from dynamic changes in the respective speed components. At every moment, however, there must be an equilibrium of forces so that the total wheel force F_Degree by the vectorial force sum of the individual wheel force vectors F_v1, F ^, F ^, F_v4 must arise, as shown in Fig. 4. In addition to the equilibrium of forces, for a state of the vehicle 1 that is stable at any moment, it is of course also necessary that the sum of all moments around a point, such as the center of gravity, must be zero. (As is known from statics, instead of equilibrium of forces, sums of the moments around 3 points can be used equivalently).

5 shows a vehicle 1, the maximum achievable (and thus available) wheel force vectors F_v1max, F_V2max, F_V3max, F_V4max per bike 2_n 2₂, 2₃, 2₄ are shown as circles or ellipses. The circles on the rear wheels show that with the rear wheel type the forces that can be transmitted in the longitudinal direction of the wheel arch and the lateral direction of the wheel arch can be the same and a wheel force vector F._Vx between zero and the maximum achievable (and thus available) wheel force vectors F_Vxmax is possible. The ellipses on the front wheels indicate that in the direction of travel of the front wheel, more braking force or acceleration force can be achieved in the direction of travel than cornering force and a wheel force vector F._Vx between zero and the maximum achievable (and thus available) wheel force vectors F_Vxmax is possible according to the ellipse. Of course, there are also other curves for the available wheel force vectors Fvimax, F_V.2max, F._V.3max, F._V.4max possible, which can be measured, for example. These curves (circles, ellipses, measured relationships) can, for example, be derived from the tire data and can be assumed to be known. The size of the curves (circles, ellipses, others) gives the maximum possible wheel force vector F._v1max, F_V2max, F_V3max, F_V4max on. The wheels 2_b 2₂, 2₃, 2₄ pressed against the road surface to different degrees, causing the 2₁ 2₂, 2₃,2₄different transferable frictional forces arise. In addition, the road conditions can be different, which can also be reflected in the size of the curve. In the example according to FIG. 5, one can see that only at the front right that wheel force vector F_V2 is achievable, which is required to generate the resulting force F_R. to compensate. The other wheel force vectors F_V1, F_V3, F_V4 can either not be reached as desired due to insufficient road grip (pressure force, road conditions) (wheels 2_b 2₄) or are not exhausted as much as possible, e.g. because of a predetermined brake force distribution (bike 2₃). The vehicle 1 will therefore get into an unwanted or uncontrollable state and may not obey the driver's wishes under certain circumstances. At this point, systems such as ABS or ESP are used in conventional vehicles 1, which attempt to remedy this driving condition by intervening individually for each wheel.

With knowledge of the currently available wheel force vectors F_v1max, F_V2max, F_V3max, F_V4max(from the known circles, ellipses or curves), however, the momentary wheel force vectors F_V1, F_V2, F_V3, F_V4 but can be combined in such a way that the current total wheel force F_Degree the resulting force F currently produced by the driving condition and the driving environment_R. cancels, as indicated in Fig.6. It is assumed that the sum of the moments around a point (e.g. the center of gravity) also becomes zero. By applying braking forces individually for each wheel, it is therefore possible for each wheel 2_b 2₂, 2₃, 2₄ a certain point of the wheel force vector curves can be set so that the wheel force vectors F_V1, F_V2, F_V3, F_V4 at each moment in the sum of the opposing resulting force F_R. are equivalent to. The individual wheel braking forces are consequently distributed according to the modeled circles and ellipses (or curves in general).

You can see very clearly here the complete difference to a normal pressure-actuated / 18

AT 12 010 U2 2011-09-15 Austrian

Patent office

Brake with ABS and ESP, because there is first left-right braking and a certain ratio at the front. Then suddenly a driving state that triggers ABS or ESP occurs and an attempt is made to achieve an improvement by locking and letting go of the wheels (ABS) or by applying wheel-specific braking forces (ESP). In contrast to pressure-actuated brake systems with ABS or With a wheel force-oriented solution, ESP is not evenly built up braking torque until the wheels lock and then the braking system switches to ABS or ESP state, but the required wheel-specific braking forces are built up and distributed according to the modeled circles and ellipses (curves) right from the start.

For this purpose, a quick observation of wheel speeds is advantageous in order to quickly identify excessive slip and thus make the wheel force vector curves dependent not only on the dynamic wheel load distribution, which is essentially specified by the brake control, but also on the actual road grip conditions. Knowledge of the instantaneous braking forces or braking torques is also required in order to be able to regulate the braking forces accordingly.

A measurement of the wheel speed that is sufficiently fast for the subject matter can be carried out with the conventional toothed disks (for example of the wheel speed sensor 5) on the wheels 2 ^ 2₂, 2₃, 2₄ be performed. Assuming, for example, a disk with 30 teeth on a wheel with a diameter of 0.64 m, this wheel has a circumference of 2 m and rotates ten times at 72 km / h. At 72 km / h, 300 teeth per second pass the meter. In the conventional method, teeth are counted in a defined gate time, this could be, for example, 1/10 s with 30 counts. Unfortunately, this is neither particularly fast (every 1/10 s) nor particularly accurate (resolution 30). On the other hand, in the method presented here, one could measure the time that lies between two teeth and thus make a very precise time measurement every 1/300 s, which enables very precise wheel speed measurement in short subsequent times. This enables wheel slip to be detected very quickly, for example by determining a wheel speed that does not match the current driving condition. Alternatively, the wheel speed could also be determined by determining the vehicle speed above ground, for example by GPS or Doppler effects.

For the individually adjustable braking forces, as are possible, for example, with electrically operated brakes, a precise control of the braking force actually achieved is necessary. This can take place, for example, via the current of the electric motor of the brake actuator, which presses the brake lining on. This is for brakes with a low or . known self-reinforcement sufficiently accurate. An additional medium-term calibration can be carried out by the (e.g. The estimated wheel braking torque is compared with the respective achieved wheel slippage via the brake actuator current and in the medium term it is assumed that similar road friction conditions exist on the wheels, i.e. in the medium term the friction brakes are operated in such a way that the same average slip is achieved in the medium term.

A very fast and inexpensive alternative measurement of the wheel braking torque (or the wheel braking forces) on a brake with a floating caliper (as is conventionally used in vehicles) can be carried out in a simple manner by an additional spring on the brake pad of a floating caliper brake remote from the brake actuator, as shown in FIG. 7 shown schematically and explained below. In a brake caliper 20, the brake pad 21 remote from the brake actuator is arranged displaceably, as is well known, for example in a recess 22 in the brake caliper 20, the recess 22 simultaneously forming the guide for the brake pad 21. A spring 23, one end of which is held on the brake caliper 20, presses the brake lining 21 in the unbraked state against the upper stop 24, which is formed here by the upper limit of the recess 22. If, in the event of braking, the drag force (braking force) F_B. becomes larger, the brake lining 21 migrates downward with compression of the spring 23 until a lower stop 26, here formed by the lower boundary of the recess 22, is reached. The path of the spring 23 between the two stops 24, 26 is a measure of the braking force F._B.. To get the braking force F_B. to determine, either the

12/18

AT 12 010 U2 2011-09-15 Austrian

Patent Office covered path of the brake lining 21 or only one point (lifting from the upper stop 24, reaching the lower stop 26) or both points can be measured. A precise relationship between brake actuator data (eg electric motor current) and braking force F_B. getting produced. One thus obtains one point (or two or more points) on a relationship between the travel of the spring 23 and the braking force F._B.. If the brake is not self-energizing, a linear curve can be assumed, for example. With a known self-amplification or a known course of the self-amplification during braking, the relationship between the travel of the spring 23 and the braking force F_B. be determined.

The driver's wishes are also advantageously incorporated into the brake control, for example a brake pedal position or force on the brake pedal, the accelerator pedal position or the steering wheel angle (as a measure of the desired curve radius). An accelerator pedal (throttle lever) generally has the same effect as a braking request with a different sign. The speed of the driver's wishes can also be evaluated. So, for example A quick release of the accelerator pedal can be used as an indication of a subsequent braking, in order to already apply the brake pads in advance (and, if necessary, to release them again if no braking follows).

In the ideal and normal case, both the desired curve radius and the desired braking or acceleration can be achieved from the wheel grip that can be achieved. If the achievable wheel grip is overstrained, the last driver request can be classified as essential, e.g. a last curve radius reduction (further steering angle) can be classified as having priority over braking because it could be an evasive maneuver. As a result, the brake control system primarily serves the wish that is classified as essential.

If the individual wheel braking forces are set as described above, the vehicle 1 theoretically remains in an unchanged stable state until the possible road grip is finally exceeded. If a certain yaw rate is introduced into the vehicle 1, this would theoretically be retained, since the braking method according to the invention inherently does not introduce any new torques (which could change the yaw rate) into the vehicle 1. The slightest deviations in the individual wheel settings in the limit state would, however, cause small moments that cause a change in the yaw rate. The actual yaw rate is therefore advantageously compared with the desired yaw rate (for example derived from the steering wheel angle) and corrected accordingly by adapting the braking forces for each wheel.