DE102008011947A1 - Robotic vehicle and driving method for a robotic vehicle - Google Patents

Abstract

The invention relates to a robot vehicle (1) designed for independent operation, with drive means for steering and moving the robot vehicle (1) and with control means for driving the drive means and with odometry sensor means (7) for acquiring odometry data. According to the invention, provision is made for compass means (8) for determining global orientation data about the global orientation of the robotic vehicle (1) and for the control means (3) to design the drive means on the basis of the global alignment data and the odometry data in such a way that the robotic vehicle (1), at least temporarily, along a preferred movement axis (11) moves. Furthermore, the invention relates to a driving method for a robot vehicle (1).

Description

State of the art

The The invention relates to a self-employed Operation trained robotic vehicle according to the The preamble of claim 1 and a driving method for driving a robotic vehicle according to the preamble of Claim 16.

at autonomous, d. H. for independent operation trained, lawn mowers is the problem that the Working surface, namely a lawn, usually not surrounded by a fixed boundary that is considered clear recognizable limit serve for the orientation of the lawn mower could. In this respect, it is compared with lawnmowers for example, home robots difficult, suitable drive solutions to realize. This problem is aggravated by the fact that lawns usually an irregular outer contour exhibit. To prevent the lawnmower from covering the lawn leaves during self-employment, it is known to limit from a border wire.

One Another disadvantage of the known autonomous lawnmower consists in that the lawn mower is within that of the current flowing through Ladder limited workspace performs a random navigation, which leads to a dirty cross-sectional image and on the other hand, no complete coverage of the mowing Area guaranteed. In addition, results from the multiple departures already mowed areas an elevated Energy consumption and damage to the turf plant.

Next The described, known robotic vehicles are robotic vehicles whose control means are the drive means based on odometry data drive. The position determination of the robot vehicle exclusively However, based on odometry data is relatively inaccurate.

An improved control system for autonomous lawnmowers is from the GB 2277152 A1 known. The known drive system includes a plurality of spaced apart landmarks defining a work surface. The autonomous lawnmower actively communicates with the landmarks to determine its position and calculate a distance based on this positional data. Such a drive system is not economically feasible. In addition, there is a need to mark the working surface permanently with landmarks.

Disclosure of the invention

Technical task

Of the Invention is based on the object, an autonomous, cost-effective Propose robotic vehicle, with the addition of or Alternatively to a random navigation, a departure strategy can be realized is. The working area should preferably be as far as possible with the robotic vehicle completely, at least largely avoiding Multiple crossings, be retractable or editable. Furthermore, the object is a correspondingly optimized driving method to propose for autonomous robotic vehicles.

Technical solution

These Task becomes regarding the robotic vehicle with the features of claim 1 and in terms of the driving method with the Characteristics of claim 16 solved. Advantageous developments The invention are specified in the subclaims. In The scope of the invention also includes all combinations at least two of in the description, the claims and / or features disclosed in the figures. To avoid repetition should also be disclosed in accordance with the device features as disclosed according to the method and claim claimable be. Likewise, according to the method disclosed Features as disclosed by the device apply and claimable.

The invention is based on the idea of enabling the robotic vehicle to travel along at least one, preferably at least approximately parallel, to a preferred travel axis in that the control means, in addition to the odometry data, provide global alignment data about the robotic vehicle's global orientation, in particular the current global travel direction of the robotic vehicle. The global orientation data in the sense of the invention does not mean coordinates in a global coordinate system, but rather the orientation of the robot vehicle, that is to say data about in which direction the robot vehicle is aligned and / or moves. In other words, for tracking and / or tracking control for maintaining a preferential direction of movement (forward or return direction) or orienting along the preferential traveling axis, in addition to the odometry data determined by the odometry (odometer), the global alignment data determined by compass means is transmitted takes into account the global orientation of the robotic vehicle. In this case, under odometry data, for example, wheel speeds or distances determined preferably from wheel speeds are to be ver stand, which may be different for a differential drive for two wheels of an axle. Preferably, the odometry data also includes steering angle information in the case of providing a pivotable steering. Taking into account odometry data, it is possible to implement a departure strategy with a comparatively low technical outlay, resulting in a more beautiful mowing pattern and, secondly, lower energy consumption and a reduction in mowing time, in the case of a robotic vehicle designed as a lawnmower robot. In the case of the formation of the robotic vehicle as a lawnmower vehicle also protects the lawn due to the avoidance of multiple overruns. Furthermore, it is possible to dispense with expensive and technically complex landmarks, which also negatively influence the aesthetic appearance of the work area, in particular of the lawn. With regard to the concrete design of the drive means there are a variety of ways. For example, it is possible to design the drive means as an airbag drive. Alternatively, it is possible to provide at least one steerable wheel. Preferred is an embodiment of the drive means as a differential drive, preferably without differential gear with at least two separate drive motors (one motor per driven wheel).

In Development of the invention is provided with advantage that the Control agent in addition to odometry data and global Alignment data the temporal alignment changes (Angular velocity) of the robotic vehicle, which are detected by means of angle sensors. In other words consider the control means in development of the invention Yaw angle change data of the robotic vehicle. At the angle sensors For example, it may be at least one rotation rate sensor act. Due to the additional consideration this temporal orientation change is even more exact Maintaining the preferred direction of movement possible.

From particular advantage is an embodiment in which the Control means the global alignment data and the temporal alignment change data to resulting alignment data, d. H. to a resulting Orientation estimates are formed fusioning. For this In particular, the use of a temporal filter, preferably a Kalman filter that produces a resultant at certain times Orientation estimate plus the corresponding Variance determined. The control means preferably control the drive means Consideration of this resulting alignment data at.

From particular advantage is an embodiment in which the merged, d. H. the resulting alignment data or the resulting orientation estimate from the tax credits be merged with the odometry data / will order this way To obtain as exact as possible position data of the robot vehicle. Position data includes position data (X, Y coordinates) of the Robot vehicle and an angle φ to understand. In which Angle φ may be that of the compass means determined global targeting data, or those from the merger resulting with the alignment change data Alignment data or preferably by merging with optimized angle estimate obtained from the odometry data act. Particularly preferred for the fusion of the resulting Registration data and the odometry data is a temporal filter, preferably a Kalman filter is used. In case that only global alignment data available stand, for example because of cost reasons on the angle sensor means dispensed with determining the relative orientation of the robotic vehicle can, in particular using at least one Kalman filters, also directly the global alignment data with the odometry data to location data or to a more exact location estimate On the basis of which the control means are the driving means then drive so that the robotic vehicle along the preferred movement axis moves. Preferably, the robot vehicle of the Controlling means controlled such that it is parallel to the preferential movement axis moves.

Preferably can for the two previously described fusing steps a common Kalman filter can be provided. additionally or alternatively, at least one of the Kalman filters or the common Kalman filters are extended so that at least a parameter such as the wheel diameter or a systematic error of the rotation rate sensor, for calibrating at least one sensor (eg the odometry sensor means and / or the angle sensor means, etc.) is appreciated.

In a development of the invention, it is advantageously provided that the position data (position data estimate) determined by the control means are determined in a local coordinate system or converted into a local coordinate system. Preferably, this coordinate system is set each time the robot vehicle is restarted and / or before a new lane or train is departed and / or before a new movement pattern is traveled, the respective starting position of the robot vehicle (X _Start , Y _Start , φ _Start ) the Location of the local coordinate system determined in the global coordinate system.

Of particular advantage is a Ausfüh tion form of the invention, in which the control means are not directly the odometry data or the global and / or relative orientation data, ie the immediate sensor information taken into account, but position data which were determined by merging a plurality of sensor data, preferably as described above. The attitude data are used by the direction correction correction means, ie for a lane keeping control, to follow the preferential travel axis, preferably to travel parallel (alternatively in serpentine or zigzagging) along it. In the simplest case, the preferred travel axis along which the control means drives the drive means is an axis which coincides with or runs parallel to the longitudinal central axis of the robotic vehicle when the robotic vehicle is set up or during the initialization of the robotic vehicle. Additionally or alternatively, the preferred movement axis can be specified manually via corresponding input means. Additionally or alternatively, an embodiment can be realized in which the control means automatically determine the preferred travel axis taking into account sensor information. For example, the preferred travel axis may be defined by a fixed work area boundary, such as a wall, such that the control means defines a preferential travel axis parallel to the boundary. Alternatively, a defined axis, for example a north / south axis, or northwest / southeast axis can be selected or specified as a preferred movement axis.

In Development of the invention is provided with advantage that the Robot vehicle or the drive means after a certain time the method along the preferential locomotion axis a steering movement perform. Preferably, the robotic vehicle is included turned by 180 °, after the rotation in the opposite direction Direction along the preferential locomotion axis (preferably parallel to this) to proceed. Alternatively, the robotic vehicle drives after the steering movement along a new, angled to the original one Preferential locomotion axis extending preferential locomotion axis. So that the control means at the right time a steering movement can initiate the drive means, for example a sensor can be provided which recognizes the correct "time". This may be, for example, an inductive sensor, the detects a buried, current-carrying conductor. additionally or alternatively, the sensor may be designed as an obstacle sensor be that of an obstacle (especially in the direction of travel) of the autonomous robot vehicle recognizes. The obstacle sensor can be tactile, capacitive, ultrasound-based, laser-based or camera-based working trained be. Likewise it is possible that the steering movement after Departing a defined route is initiated, for this purpose the odometry data can be used. additionally or alternatively, the decision whether a Lenkbewe movement, in particular a turnaround to be made, based on it whether from a sensor, for example an ultrasonic sensor It is determined whether the autonomous robotic vehicle is still on the work area to be edited, or already edited Workspace is located. In the case of a lawnmower can for example, via the measurement of the motor current or on be detected in another way, whether the robotic vehicle is on an already mowed area or on a still mowing Area is located. Alternatively, the steering movement can then be initiated when an appropriate sensor is leaving the work area (eg lawn area) or threatening to leave the work area recognizes.

From particular advantage is an embodiment in which the Control means drive the drive means such that the robotic vehicle meandering on the work area moves. For this purpose, the control means control the drive means such that the robotic vehicle, especially when reaching the limits of Work area reverses and then again moves in the opposite direction along the preferential movement axis, preferably parallel to the previously traveled lane, preferably such that the robot vehicle slightly overlaps the previously worn track, in case of the formation of the robotic vehicle as a lawnmower vehicle a clean cut or a completely edited To receive area.

in the With regard to the concrete training of the compass means there are different Options. For example, it is possible the compass means as a magnetic compass and / or polarization filter compass train. However, an embodiment is particularly preferred the compass means in which this satellite navigation receiver, in particular a GPS receiver, preferably with integral logic unit include.

It is particularly preferred if the compass means determine the orientation of the robot vehicle on the basis of a three-dimensional velocity measurement by utilizing the Doppler effect of at least one carrier signal, wherein the specific orientation can be calculated via trigonometric functions from the three-dimensional vector velocity determined by the satellite navigation receiver. This functionality of determining the three-dimensional speed is already included in known satellite navigation receivers - it only needs From the three-dimensional speed, for example, the global orientation of the robot vehicle can be determined via trigonometric functions. This task can either be taken over directly by the satellite navigation receiver or alternatively by the control means.

From particular advantage is an embodiment in which the Angle sensor means for detecting the relative or local orientation of the robotic vehicle at least one rotation rate sensor for determining a time-yaw angle change (Yaw angle change) include.

The Orientation of the robot vehicle by the control means can be further improved by adding the angle of inclination of the robotic vehicle is detected. The tilt angle data is preferred of the robot vehicle taken into account in the estimation of the position data, for example, by the inclination angle data, in particular by means of a Kalman filter, with the odometry data and / or the global Alignment data and / or the orientation change data and / or the resulting alignment data.

Farther It is possible, in addition or as an alternative to considering inclination angle data global position data to consider for driving the drive means. These global position data (eg X, Y coordinates) can be for example with the aid of a satellite navigation receiver at the same time the global alignment data determined. The global position data can, in particular by using a Kalman filter, with the odometry data and / or the global alignment data and / or resulting alignment data be merged.

From It is particularly advantageous if it is the robotic vehicle is a gardening vehicle. Preferably, the robot vehicle is as Lawn mower vehicle with a mower, whereby by the consideration of Odometriedaten and at least of global alignment data the realization of one directional mowing of the working area (lawn) becomes.

The Invention also leads to a driving method for driving of trained for driving and moving a robotic vehicle Drive means taking into account odometry data. The core of the invention is, in addition to the Odometriedaten for driving the drive means global alignment data via to consider the global orientation of the robotic vehicle. It is particularly preferred, in addition to the global alignment data Orientation change data on the temporal change the orientation of the robot vehicle for driving the drive means to take into account. Advantageous embodiments of this Driving method are directly from the previous description preferred embodiments of an autonomous robotic vehicle derivable, in which such a driving method is realized.

Brief description of the drawings

Further Advantages, features and details of the invention will become apparent the following descriptions of a preferred embodiment as well as from the drawings. These show in:

1 FIG. 2 a schematic representation of an autonomous robot vehicle designed as a lawnmower vehicle, FIG.

2 the location of a local X'-Y 'coordinate system in a global XY coordinate system (WGS84),

3 a possible departure strategy with a meandering trajectory, and

4 a schematic representation of a data fusion.

Embodiments of the invention

In The figures are the same components and components with the same Function marked with the same reference numerals.

In 1 is an autonomous, ie trained for self-service robotic vehicle 1 shown. The robot vehicle 1 is as a lawn mower vehicle with a merely schematically indicated mowing tool 2 educated. Actually known drive means for steering and moving the robot vehicle 1 are not shown for clarity. The drive means comprise per driven wheel an electric motor, ie a differential drive for driving and steering the robot vehicle 1 , To operate the mowing tool 2 An additional motor is used. The robot vehicle 1 further comprises control means 3 (Arithmetic unit), which are drivingly designed and arranged the drive means. The control means 3 are signal-conducting with a location sensor 4 connected in the embodiment shown in the middle between two driven wheels 5 . 6 is arranged. The location sensor 4 includes odometry sensor means 7 for acquiring odometry data by monitoring the rotational speeds of the wheels 5 . 6 , Furthermore, the location sensor system includes 4 compass means 8th for determining global orientation data about the global orientation of the robotic vehicle 1 , The compass means 8th include in the embodiment shown Example, a satellite navigation receiver with an integral logic unit. In the specific embodiment, a GPS receiver is used for this purpose. The GPS receiver is designed such that it has the global orientation or an estimate for the global orientation of the robotic vehicle 1 based on the measurement of three-dimensional velocities of the robotic vehicle 1 determined in a global coordinate system (here WGS 84). The 3-D speeds of the robot vehicle 1 be determined in a conventional manner, taking advantage of the Doppler effect, the carrier frequency of the satellite signal depending on the speed of the robot vehicle Velocity 1 changed. It should be noted that the global orientation, ie the global orientation of the robotic vehicle 1 in the embodiment shown, due to the illustrated measuring principle only during movement of the robot vehicle 1 can be determined with a minimum speed.

Furthermore, the location sensor includes 4 Angle sensor means 9 (Here, a rotation rate sensor) for determining orientation change data about the orientation change of the robot vehicle 1 , With the help of the angle sensor means 9 is the yaw angle (yaw angle) of the robotic vehicle 1 , ie the angle around a normal to the surface extension of the robotic vehicle 1 determinable.

The location sensor 4 ie both the odometry sensor means 7 , the compass means 8th as well as the angle sensor means 9 are over a data cable 10 signal-conducting with the control means 3 connected, which evaluate the sensor signals, or as will be explained later, suitably link, such that the robotic vehicle 1 parallel to the preferential locomotion axis 11 emotional.

The control means 3 In a first step, the global alignment data of the compass means are merged 8th (here the GPS receiver) with the orientation change data of the angle sensor means 9 (here the rotation rate sensor) to an estimate of the global orientation, ie resulting alignment data. In a next step, these merged alignment data will then be compared with the odometry data of the odometry sensor 7 fused to an estimate of location data. Algorithmically, in the embodiment shown, the procedure is as follows: First, the sensor information of the compass means 8th and the angle sensor means 9 fused by means of a Kalman filter. The result of this fusion, ie the resulting alignment data (resulting estimate), is then fused to position data (X, Y, φ) (position data etch) using a Kalman filter (preferably the same Kalman filter). At the same time the robotic vehicle drives 1 to initialize the state straight on until the first reliable global alignment data of the compass means 8th to be obtained. The tax credits 3 determined position data are used as actual values of a tracking control, which has the goal of the robotic vehicle 1 parallel to the preferential locomotion axis 11 to move. Based on the tracking control is in the embodiment shown in 3 shown meandering movement pattern 12 of the robotic vehicle 1 on a workspace 13 realized.

For the tracking control, a local coordinate system is initially used during initialization 14 defined, whose origin 15 through the starting position (X _start , Y _start , φ _start ) at the start of the robot vehicle 1 is determined. The location of the robotic vehicle 1 in this local coordinate system 14 is given below by the coordinates X 'and Y' and by the orientation φ '. In this local coordinate system 14 runs the preferential locomotion axis 11 parallel to the X'-axis or it corresponds to this. Where φ 'is the orientation of the robotic vehicle 1 relative to the preferential locomotion axis 11 , The position data is acquired during tracking by the controller 3 in the local coordinate system 14 Converted so that X 'of the already traveled route along the preferential locomotion axis 11 and Y 'of the deviation from the lane to be maintained along the preferential travel axis 11 and φ 'is the angular deviation of the track to be maintained and the preferred travel axis 11 equivalent. A control (preferably optimal control) of the control means 3 now controls the drive means (differential drive) so that the lane deviation Y 'and the angular deviation φ' are minimized.

This in 3 shown movement pattern 12 "Meandering" is accomplished by making a 180 ° turn at the end of each of the parallel tracks and a parallel track in the opposite direction from the robotic vehicle 1 is traversed. The decision as to when a turn should be made can be made dependent on different sensors and / or conditions. For example, it is possible for a sensor to detect a surround wire that is the work area 13 limited, whereupon a turn is made. Alternatively, an obstacle sensor detects an obstacle in the direction of travel of the robotic vehicle 1 , The obstacle sensor can be designed to be tactile, capacitive, ultrasound-based, laser-based or camera-based. Alternatively, a turn is made when a maximum distance has been traveled along the track X '(limitation of the maximum track length). Alternatively, a sensor detects that the robotic vehicle 1 already on the mowed area or alternatively outside the lawn or in the area of the lawn boundary. This can be done by monitoring the Motor current of the mowing tool 2 be determined.

In 4 schematically is a possible data flow in a robotic vehicle 1 shown in a preferred embodiment. You can see the location sensor 4 with their compass means 8th and the angle sensor means 9 , The compass means 8th , here a GPS receiver, obtain global alignment data φ _GPS on the global orientation of the robotic vehicle. The angle sensor means 9 comprise a yaw rate sensor which detects angular velocities φ.

The angular velocity φ and the global orientation data φ _GPS become a first Kalman filter 17 fed. This calculates a merged alignment estimate with variance, thus resulting alignment data φ, which in turn is a second Kalman filter 18 be supplied. These are used by odometry sensors 7 determined odometry data, here the distance traveled by a left driven wheel s _L and the distance traveled by a right driven wheel s _R fed. The second Kalman filter calculates therefrom fused position data X, Y, φ on the basis of which the control means drive the drive means of the robot vehicle, wherein the control means from the global position data X, Y, φ the position data X ', Y', φ 'in the local coordinate system 14 calculated as described below using a preferred algorithm.

The following describes a preferred algorithm for tracing parallel lanes (lanes). It is assumed that the start takes place at the time t and the priority movement axis 11 the current robot orientation φ corresponds.

1. Initialization

First, the system is initialized. Here, the position (X, Y, φ) at time t becomes the output of the second Kalman filter 18 saved. The stored position defines a local coordinate system 14 as exemplified in 2 is reproduced. The position (X ', Y', φ ') is calculated in the following by the projection of the output data of the second Kalman filter 18 in this local coordinate system 14 ,

2. Lane Keeping Control

• 2.1 Es werden laufend aktuelle Lagedaten (X', Y', φ') berechnet, wobei X' der gefahrenen Strecke entlang der Vorzugs-Bewegungsachse 11, Y' der lateralen Abweichung von der Vorzugs-Fortbewegungsachse 11 und φ' dem Ausrichtungsfehler in Bezug auf die Vorzugs-Fortbewegungsachse 11 entspricht.
• 2.2 es werden laufend Antriebsstellkommandos mittels eines Reglers generiert, der Y' und φ' zu minimieren sucht. Vorzugsweise kommt dabei ein Optimalregler zum Einsatz, bei dem gleichzeitig der Energieverbrauch minimiert werden kann.

As long as no end of track (end of track) is detected, proceed as follows:

• 2.1 Current position data (X ', Y', φ ') are continuously calculated, where X' is the distance traveled along the preferred motion axis 11 , Y 'of the lateral deviation from the preferential locomotion axis 11 and φ 'is the registration error with respect to the preferred travel axis 11 equivalent.
• 2.2 Drive command commands are continuously generated by means of a controller which seeks to minimize Y 'and φ'. Preferably, an optimum controller is used, in which at the same time the energy consumption can be minimized.

3rd end of track

As soon as A track end is detected becomes a turning maneuver with trajectory misalignment carried out. The algorithm will be aborted if that Maneuver is not successful.

4. Calculation of a new one local coordinate system

• 5. Daraufhin wird erneut Schritt 2. ausgeführt.

It will be a new local coordinate system 14 which is rotated by 180 ° and the path offset (track offset) in the Y-axis compared to the previous local coordinate system 14 is moved.

• 5. Step 2 will be executed again.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - GB 2277152 A1 [0005]

Claims (17)

Self-contained robotic vehicle ( 1 ), with drive means for steering and moving the robot vehicle ( 1 ) and with tax funds ( 3 ) for driving the drive means and with odometry sensor means ( 7 ) for acquiring odometry data, characterized in that compass means ( 8th ) for determining global orientation data about the global orientation of the robotic vehicle ( 1 ) and that the control means ( 3 ) the drive means is designed on the basis of the global orientation data and the odometry data in such a way that the robot vehicle ( 1 ), at least temporarily, along a preferential locomotion axis ( 11 ). Robot vehicle according to claim 1, characterized in that angle sensor means ( 9 ) for, in particular simultaneously, determining alignment change data on the temporal change of the orientation of the robotic vehicle ( 1 ) and that the control means ( 3 ) the relative alignment data for driving the drive means are designed to take into account. Robot vehicle according to claim 2, characterized in that the control means ( 3 ) the global and the relative orientation data, in particular by means of a Kalman filter ( 17 ), merging into resulting alignment data. Robot vehicle according to claim 3, characterized in that the control means ( 3 ) Position data of the robotic vehicle ( 1 by fusing the resulting alignment data with the odometry data, in particular by means of a Kalmen filter ( 18 ), determining are formed. Robot vehicle according to claim 4, characterized in that the control means ( 3 ) the location data in a local coordinate system ( 14 ) are formed determinative. Robot vehicle according to one of claims 4 or 5, characterized in that the control means ( 3 ) the position data are designed as actual values for a tracking control. Robot vehicle according to one of the preceding claims, characterized in that the preferential locomotion axis ( 11 ) manually specifiable and / or by the control means ( 3 ), in particular taking into account sensor information, is automatically determined. Robot vehicle according to one of the preceding claims, characterized in that the control means ( 3 ), preferably in dependence on sensory means determined by sensor means, the drive means are designed to be such that, if at least one condition is met, execute a steering movement and preferably the robotic vehicle ( 1 ) turn it over. Robot vehicle according to claim 8, characterized in that the control means ( 3 ) the drive means are designed in such a way as a function of the position data such that the robot vehicle ( 1 ) after the steering movement along the preferential locomotion axis ( 11 ) parallel to a traced before the steering movement track, preferably the previously worn track laterally overlapping, moved. Robot vehicle according to one of the preceding claims, characterized in that the compass means ( 8th ) comprise a satellite navigation receiver, in particular a GPS receiver, and / or a magnetic compass and / or a polarization filter compass. Robot vehicle according to claim 10, characterized in that that the satellite navigation receiver is the global alignment data taking into account speed measurements, in particular taking advantage of the Doppler effect of at least one carrier signal, is designed determining. Robot vehicle according to one of claims 2 to 11, characterized in that the angle sensor means ( 9 ) a yaw rate sensor for determining a time yaw angle change of the robot vehicle ( 1 ). Robot vehicle according to one of the preceding claims, characterized in that inclination angle measuring means for determining inclination angle data of the robotic vehicle ( 1 ) and that the control means ( 3 ) the inclination angle data for driving the drive means are designed to take into account. Robot vehicle according to one of claims 10 to 13, characterized in that the control means ( 3 ) by means of the satellite navigation receiver satellite navigation position data for driving the drive means ( 3 ) are designed to take into account. Robot vehicle according to one of the preceding claims, characterized in that the robot vehicle ( 1 ) as a gardening vehicle, in particular as a lawnmower vehicle with a mowing tool ( 2 ), is trained. Control method for driving, from to Steering and moving a robot vehicle ( 1 ) in consideration of odometry data, characterized in that global alignment data about the global orientation of the robotic vehicle ( 1 ) and taken into account for driving the drive means. A driving method according to claim 16, characterized in that alignment change data on the temporal change of the orientation of the robot vehicle ( 1 ) and taken into account for driving the drive means.