DE102014221763A1 - A method of automatically controlling objects within a confined area intended for the manufacture or maintenance or parking of a vehicle - Google Patents

Abstract

The invention relates to a method for automatically controlling one or more objects (41, 42, 43) within a spatially delimited area (1), which is provided or designed for the manufacture or maintenance or the parking of a vehicle (10, 20) the vehicle (10, 20) comprises a vehicle sensor system (11, 21) with at least one sensor (31a, 32a, 33a, 34a, 35a, 36a) for detecting a part of the environment of the spatially delimited area (1). The vehicle sensor system (11, 21) detects a part of the surroundings of the vehicle (10, 20) within the spatially delimited area (1) and provides first data which the object or objects (41, 42, 43) in the environment of the vehicle (10, 20). Control data for controlling the movement behavior of the object or objects (41, 42, 43) is determined by processing the first data provided by the vehicle sensor system (11, 21). There is an activation of the object or objects (41, 42, 43) as a function of the determined control data which leads to a movement or to a change of the movement or a change of the state or behavior of the object or objects (41, 42, 43). leads.

Description

The invention relates to a method for the automatic control of one or more objects within a spatially delimited area, which is provided or designed for the manufacture or maintenance or the parking of a vehicle.

Modern vehicles are increasingly equipped with a vehicle sensor system, which serve to capture their environment. As a rule, the vehicle sensor system is part of a driver assistance system or a safety system. Such vehicle sensors, which include, for example, one or more cameras, a radar or ultrasound system, a stereo camera or a light-emitting sensor system (eg laser sensor system), specializes in interaction with corresponding computing units of the vehicle, other vehicles , Persons, such as As pedestrians, or lines on a roadway on which the vehicle moves to detect. Depending on the configuration of the driver assistance system or the driving safety system, after evaluation of the detected information, a warning for the driver and / or a control of one or more actuators takes place. Due to the increasing performance of the vehicle-mounted computing unit and vehicle-integrated vehicle sensor systems, the detection and speed of detection of other objects is becoming better and better. The use of the vehicle sensors and their computing units takes place as soon as the vehicle has been transferred by its user to its intended use.

It is an object of the present invention to specify a method which enables or improves the control of objects by using a vehicle sensor system.

This object is achieved by a method according to the features of claim 1, a computer program product according to the features of claim 18 and a device according to the features of claim 19. Advantageous embodiments are specified in the dependent claims.

The invention relates to a method for automatically controlling one or more objects within a spatially delimited area provided or designed for the manufacture or maintenance or storage of a vehicle, the vehicle having a vehicle sensor system with at least one sensor for detecting part of the vehicle Surrounding the spatially delimited area includes. The vehicle sensor system detects a part of the surroundings of the vehicle within the spatially delimited area and provides first data representing the object or objects in the vicinity of the vehicle. Control data for controlling the movement behavior of the object or objects is determined while processing the first data provided by the vehicle sensor system. Subsequently, a control of the object or objects takes place as a function of the determined control data, which leads to a movement or to a change in the state or behavior of the object or objects.

The vehicle is preferably a motor vehicle, in particular an electric or hybrid vehicle.

Control data may include a desired longitudinal acceleration of the object or objects representative of driving or braking (as indicated) and at least one lateral dynamic value representative of a lateral directional change. The control data may also include vertical dynamics, motion stabilization, roll, and possibly drift. The movement possibilities of the object or objects which can be achieved by means of the control data can be taken, for example, from predetermined maneuver elements which are automatically retrieved and executed on the basis of the determined control data.

Vehicle sensors are to be understood in particular as components of a driver assistance system. Such a vehicle sensor system may include one or more cameras and / or a radar system and / or an ultrasound system and / or a stereo camera and / or a light-emitting sensor system, such as a light-emitting sensor. B. a laser sensor, in particular a laser scanner include. The vehicle sensor is used in particular for Umweisfassung. This means that the sensor (s) of the vehicle sensor system detect, measure and / or detect objects which are located in the surroundings of the vehicle. A vehicle sensor system is also to be understood as meaning those components which are designed to image the environment in a spatial reference data model.

The vehicle sensor system is coupled to at least one computing unit. The arithmetic unit can be, for example, a vehicle control unit (Electronic Control Unit, ECU). The arithmetic unit can be installed together or separately with the vehicle sensors.

Sensors in the context of this description are preferably so-called perceptive sensors. Sensors can z. As cameras with or without a static or dynamic lighting, radars, ultrasonic sensors, lidars, LED sensors, etc. be. It can also be sensors that rely on a transmission and detection of a static or dynamic, based on structured light. In this case, a predetermined light pattern, in particular in the infrared spectrum, can be emitted in a spatial area and received and evaluated with a light sensor or image sensor. Such a type of sensor is very well suited for a defined spatial area. Also, objects can be detected particularly reliably.

In particular, the vehicle sensor system comprises at least one camera and at least one radar sensor or a radar device. Particularly preferred are the sensors, which is designed to operate a driver assistance system of the vehicle.

As a result of the vehicle sensor system, additional or additional data which are preferably read out by means of RFID (= Radio Frequency Identification) can be used as part of the method. For example, a vehicle can read out a plurality of RFID transponders mounted in the spatially delimited area by means of an RFID reader. In this case, the vehicle can determine its own position and / or orientation with respect to certain points in space or objects. The at least partially automated movement of the object or objects can be done in dependence on this data. Similarly, radio wave transmitters, z. B. WLAN transmitter or Bluetooth transmitter, in particular to a direction-dependent transmission of data are used.

In the present description, an object to be controlled is to be understood as a robot, a forklift, a transport or loading device, in particular for motor vehicles. Objects may also be other motor vehicles which, for example, together with the vehicle to be executed, z. B. on a tape, manufactured or otherwise maintained. However, objects should also be understood as meaning persons who, for example, are located in the spatially delimited area which is detected by the vehicle sensor system of the vehicle, for example for a processing step or also accidentally. In particular, the object is a device which is designed for the further transport of the vehicle, in particular for turning or loading the vehicle. Such a device may include appropriate means, e.g. As a lifting device or a device for coupling a vehicle include. A lifting device may be a device that leads or grips under the vehicle and is configured to the vehicle, for. B. to rotate about a vertical axis, or laterally, substantially perpendicular to the direction of travel of the vehicle to convey.

The invention is based on the idea of utilizing the vehicle sensor system, which is installed in a vehicle to provide a driver assistance system, during the vehicle's stay in the spatially delimited area. In this case, it is provided according to the invention to control other objects as a function of the first data of the vehicle sensor system. This allows the movement and / or coordination of objects in the spatially delimited area as well as a control of their behavior.

The spatially delimited area is in particular a factory site, z. B. a hall in which the vehicle is made. The spatially delimited area also includes routes between individual buildings or a parking garage in which vehicles are parked logistically. A spatially delimited area can also be a parking garage, a carwash or another environment in which an interaction between a vehicle and one or more controllable objects is possible. It can be ensured in the spatially delimited area that only objects of a certain type, in particular predominantly objects controllable by the method, are located therein. Thus, the method is even easier to implement.

The spatially delineated area may be a specially equipped parking garage, a factory premises for a fully or partially automatic production of the vehicles or a special shunting area for an automatic maneuvering of the vehicles. In particular, the spatially delimited area is a substantially flat, in at least two dimensions extended area, d. H. no lane or road, but rather a driving surface. For example, the spatially delimited area may be a floor of a parking garage, an area for maneuvering the vehicles, a ferry or a ship for transporting motor vehicles. In such a spatially delimited area can advantageously be realized an automatic maneuvering including collision avoidance around the vehicle.

The spatially delimited area can be a completely or partially spatially closed and / or considered area. Here, the special and / or substantially predefined, preferably constant, conditions for operating the method can be created in a simple manner. For example, such a region may have a predefined illumination, so that the detection of first and / or second data by the respective sensor systems can take place in the best possible way. Also, additional means may be provided which contribute to a particularly effective and stable sensor operation. For example, objects or room parts can be provided with markings that are easily and reliably detected by the sensors can be. For example, special markings may be provided within the spatially delimited area that reflect light and / or radar waves in a predetermined manner.

Such markings can be a machine-readable code that can be read or recorded with the aid of the vehicle sensor system, eg. B. a QR code include. The code may include, for example, information about a position or about a place. The markings may be configured to reflect light and / or radar waves emitted by a sensor in a predetermined manner, the reflected pattern being distinguishable from other reflections. Likewise, the object or objects may be provided with markings. This supports efficient execution of the proposed method.

Should z. B. an interfering object occur in the spatially delimited area, it can be coordinated with the object or objects with appropriate provision of the control data in dependence on the first data a right of way. As a result, irregularities can be remedied or desired states can be established, so that dangerous situations which possibly endanger the vehicle or disrupt the operating procedure can be avoided. As a result, it is possible to increase the automation of the movement of the objects in a spatially delimited area in which, in addition to one or more vehicles, one or more objects are also controlled.

If in this context of a vehicle is mentioned, it may be a finished manufactured vehicle, but also to a only partially finished manufactured vehicle. For example, the vehicle may still move on a production line (belt) or be moved by another movement device. In this case, the moving device that moves the vehicle can also be controlled using the vehicle sensor system with regard to its state and its movement.

The first data provided by the vehicle sensor system can be image or measurement data. By means of a corresponding arithmetic unit, a measurement and / or culture formation of one or more objects and / or the object recognition as such and / or the prediction of the movement of the object can be determined from the first acquired data.

Representing one or more objects in the environment of the vehicle relates to the position and / or motion data and / or a property and / or a state of the object or objects. An object is also a part of an object. An object can be represented by a side of the object facing the respective sensor or a part of the object which is only partially located in the detection area of the vehicle sensor system.

The position may include coordinates, orientation, etc. of at least a portion of the object. The motion data may include a speed or acceleration of the object. The property may include an object class of the object (human, moving object, immovable object, vehicle of a certain class, etc.). The state of the object can, for. B. an active or passive state of the object include. The object may, for example, be part of a movable device.

The first data may represent an absolute position of one or more objects and / or a relative position of one or more objects in the vehicle coordinates. An absolute position may refer to a stationary coordinate system, e.g. B. also refer to a global coordinate system.

The first data may represent several parts of an object, for example by a so-called point cloud, a location-discrete function, an interpolated function, etc. In particular, the first data may include the information about physical boundaries of the object or objects, for example, positions of surface boundaries. In this case, the first data can represent the information about several parts of an object, in particular the object boundaries.

The data may also include interpreted information about the parts of the object, e.g. B. a movable or immovable part of the object. The first data may also include information about moving and immovable parts of an object.

In an expedient embodiment, the spatially delimited area comprises an exterior sensor system with at least one sensor for detecting a part of the environment of the spatially delimited area, the outside sensor system providing second data relating to the vehicle and / or one or more objects in the surroundings represent the vehicle.

An external sensor system may be a stationary or movable sensor system in or at the spatially delimited area. The exterior sensor system is designed to detect the vehicle and / or the surroundings of the vehicle from outside the vehicle. The external sensor system is preferably a processor unit or arithmetic unit which is configured to generate a data model from the spatial reference environment and to execute corresponding methods. The external sensor system may comprise, for example, light-emitting sensors, in particular a laser sensor system (eg comprising one or more laser scanners), in particular with corresponding data processing. The outdoor sensor may also include one or more cameras and the like. The arithmetic unit can be a conventional computer.

The second data may represent an absolute position of one or more objects and / or a relative position of one or more objects in a particular coordinate system dependent on the position of the outside sensor. Incidentally, the second data may represent parts of an object and / or one or more vehicles corresponding to the first data. In particular, the data may represent multiple parts of an object, such as a point cloud, a discrete location function, an interpolated function, etc. The second data may include information about the physical boundaries of the object (s) and / or vehicle (s), such as surface boundary locations. In this case, the second data can represent the information about several parts of an object, in particular the object boundaries. The second data may also include information about moving parts of an object, in particular in an explicit form.

It is expedient if at least part of the vehicle is represented in the second data by at least part of the contour of the vehicle, in particular from the perspective of the external sensor system. The second data expediently comprise at least one contour of another object.

The combination of the first and second data results, inter alia, in the advantage that the first data and the second data are each generated from sensor data that has been acquired from different spatial perspectives. Thus, z. For example, it is also possible to solve a problem known from the prior art in that obscurations of an object by another object or parts of the vehicle lead to incomplete or erroneous detection of the environment or to imprecise or erroneous prediction. Also conceivable consequences of a fault or a failure of the vehicle sensor system can be significantly mitigated.

In addition, there is the advantage that the spatially delimited area does not have to be designed with an elaborate sensor system, only in order to automatically control movements and / or the behavior of objects within the spatially delineated area. Thus, the possibility of an automated movement of the object or objects within the spatially delimited area can be improved by using the ever-improving vehicle sensor technology. It can be provided that the arithmetic unit or control unit of the object or objects for receiving the first data or data, which were generated as a function of the first data, is configured.

It is expedient if the external sensor system and the vehicle sensor system comprise different sensors, in particular sensors which are based on different physical principles or detect different properties of the objects. Preference is given to using combinations of physical sensor technology principles which have their respective advantages, preferably based on the respective detection perspective. In this case, the detected first data of the vehicle sensor system and / or the second data of the external sensor system can each be assigned information about a respective perspective.

A further advantage of this design variant is that with per se not safe or high-quality sensors, which can detect only limited geometric area, almost complete data can be generated, which are optimized to carry out an at least partially automated movement of the object or objects. It is possible to take into account objects and vehicles which are located in a geometric region, which can not be detected by the sensors, relative to the vehicle and / or object. For example, such areas concern an ultralight area of the vehicle and / or the object. A complete sensing of such areas is not possible according to the prior art due to a minimum measurable distance of the respective sensors, opening angles of the sensors and the design-related contours of the vehicle and / or object.

The at least partially automated movement of the object or objects to be controlled takes place, for example, with means of the object or objects. For example, for this purpose, an electric drive of the object can be controlled. Possibly. can also be provided at least partially automated movement of the vehicle. This is then done by an electric drive of an electric or hybrid vehicle or a combustion drive of a conventionally operated vehicle. The movement of the vehicle and / or object can also take place at least partially by means of a drive which is located outside the vehicle or object and is not part of the vehicle or object.

In accordance with an expedient embodiment, an activation of one of the actuators to achieve a predefined interaction with one or more of the objects is carried out. Particularly preferably, the control data is determined in such a way that the resulting activation of the object (s) leads to a predetermined, desired interaction between the object and the vehicle. Such an interaction may include a physical impact, particularly at a predetermined, desired level.

The interaction affects the object or objects, in particular objects of a particular type or a specific object. Such a desired interaction may e.g. These include, for example, a charging process in which the vehicle is loaded, the loading and / or unloading of a load of the vehicle, the retraction or engagement of the vehicle wheels in a device provided for this purpose, eg. B. a trough, fixing device, etc.

A predetermined interaction may be characterized by predetermined qualitative and / or quantitative criteria. This can be z. As a positioning of the vehicle with a predetermined orientation and a distance from the object or the measure of a physical force that may be either reached or not exceeded include.

In addition, the method allows effective prevention of undesired interaction with an object, eg. B. to avoid a collision that can lead to damage to the vehicle and / or object, or to avoid mutual blocking, in which at least one object and / or vehicle is at least temporarily hindered in its further movement.

According to a further expedient embodiment, the first data comprise at least one object constellation of in each case at least two objects, and the second data represent an object constellation of the vehicles of at least one of the objects.

Preferably, the first data and / or the second data represent an object constellation, i. H. an arrangement of objects. An object constellation can be represented by means of qualitative and quantitative data. A useful example of qualitative data is data representing at least partial obscuration of an object by another object from a particular perspective. Quantitative data may preferably represent the object positions, including the orientation of the objects to each other or with respect to a coordinate system. Particularly preferably, an object constellation, based on at least one specific time or several specific times, is represented. In this case, the first data and / or the second data may represent the same object constellation for at least two different points in time. The first data and / or the second data may also include the information at these times. Thus, a dynamic development of an object constellation can be better evaluated in further steps of the method.

The first and / or second data may further represent an object constellation each comprising at least two objects. Thus, there is the advantage that in the control of the movement of the vehicle and distances between at least two objects from different directions can be easily considered. It is advantageous here if the first and the second data objects represent constellations, wherein two or more objects in the two constellations match. So z. B. also for different parts of a movement process of the object or properties relevant properties of an object constellation can be detected and taken into account. For example, a relative speed and relative acceleration between certain parts of an object or object contours and the vehicle or of vehicle contours can be detected.

The second data may include information about a relative movement and / or relative acceleration of specific parts of at least one object or object contours and of the vehicle or vehicle contours.

The control data can be determined very easily because the distances between the vehicle and the object or objects can be removed if necessary without high computing power. Depending on such a constellation, the determination of the control data can be carried out effectively and essentially in real time.

In this case, the method according to this refinement can be designed for determining a plurality of collision risks, which relate to different locations of a vehicle contour and / or object contour. Control data may be generated such that the movement of the object or objects is optimized to the least possible risk and / or a minimum number of maneuvers and / or a minimum time to reach an (intermediate target).

The vehicle sensor system and / or the external sensor system can detect an object or an object constellation in at least two dimensions. For example, two dimensions in which a camera can well represent an object or object constellation are perpendicular to the axis between Camera and object. In this case, the two dimensions of the external sensor can be chosen such that they differ substantially, at least substantially and in the most possible spatial arrangements of the objects of the two dimensions of the vehicle sensors.

According to a further expedient embodiment, the detection of the surroundings of the vehicle by the vehicle sensor system and by the external sensor system takes place in a temporal dependence, in particular synchronized. This results in the advantage that the resulting first and second data relate essentially to the same points in time. As a result, a comparison of the first and second data can be carried out very easily. Such synchronization can be wireless. It can be z. B. the vehicle sensor control a detection time of the external sensor and / or the external sensor control the detection time of the vehicle sensor and / or a common clock the detection time of the vehicle sensors and outdoor sensors.

According to a further expedient embodiment, the control data is determined as a function of the first data and the second data, wherein third data are determined and provided from a comparison of the first data and the second data, which relate to the same object or the same object constellation. The third data can be determined from the first and second data by a special data fusion adapted to the method. It can from the comparison of the data, eg. B. by a first object hidden parts of a second object and / or movements of the second object, the third data are determined. It is also possible to determine distances between the regions of two objects that partially overlap (from the perspective of the vehicle sensor system and / or an external sensor system).

The comparison of the data may relate to spatial and / or temporal criteria and / or a detection perspective and / or a probability of statement.

Third data may be generated that includes a more complete representation of objects or object constellations and / or multiple motion data in a particular spatial area than the first data or the second data taken as such. The movement data may include the object movements within the object constellation, in particular as relative speeds or relative accelerations of the objects or object parts, and areas of the vehicle.

When comparing the first data and the second data, it is possible to generate the third data in which the incomplete or inaccurately mapped first data is replaced by the second data, which are essentially related to the corresponding position and time. Thus, there is also the advantage that z. B. dependent on the perspective error (eg, a reflection of electromagnetic waves, a glare of a camera by a strong light source from an unfavorable direction) to no breakage of the semi-automatic movement of the vehicle or lead to a risk of error.

The third data may in this case include information about the correspondence and / or the non-conformity of the first data and the second data with respect to an object or an object constellation.

According to a further expedient embodiment, the respective detection perspective of an object is taken into account by the external sensor system and / or by the vehicle sensor system during the adjustment of the first data and the second data. In this case, further, an object constellation related data, for. B. by triangulation, are determined. It may also be a detection of the presence of an object and an assignment of the at least one first data and the at least one second data to the object. In particular, at least two perspectives of at least two sensors of the external sensor system can be taken into account. This means that the object or objects are moved in such a way that the view of currently relevant object constellations from several directions is taken into account.

In the adjustment of the first data and the one or more second data, the information about the detection perspective of the vehicle sensor system and the external sensor system can be used to determine the position and / or the property of the object with a higher speed and / or probability that these are correct or incorrect. This allows more accurate or reliable data to be generated.

In accordance with a further refinement, by comparing the first data determined with the respective vehicle sensor systems of a plurality of vehicles, control data can be obtained by which an at least partially automatic change in the control of the object or objects can be achieved by generating learning data. The learning capability of the method provided thereby enables a better and more accurate response to the occurrence of changes. In particular, in this case, sensing methods and / or object recognition methods and / or trajectory planning methods can be improved by the evaluation of a large number of data from the past.

According to a further expedient embodiment, the exterior sensor is arranged above and / or below a part of the area in which the vehicle is moved, so that the vehicle is sensed from a top view and / or bottom view of straight or obliquely above or below , Contours from this perspective are particularly meaningful in terms of an evaluation. By a spatially favorable position of the external sensor, in particular substantially from above the vehicle, collision protection can be ensured to all or most lateral vehicle surfaces.

In accordance with a further expedient refinement, at least part of the determination of the control data, in particular the determination of a trajectory of the object (s) and / or space regions of the spatially delimited area not passable by the object (s), takes place in a processor or arithmetic unit outside the vehicle. In the spatially delimited area, wireless communication between the vehicle and the remote computing unit may occur at a much higher data rate than is the case on public roads or squares. Therefore, data fusion or data networking can also take place in the case of partially processed data, which typically involves larger amounts of data. This allows data fusion to be performed at a higher performance compared to a fusion of non-pooled data. In addition, the use of a stationary computing unit allows a cheaper computing power. This is particularly advantageous in combination with the powerful vehicle sensor system, which detects the environment from the current vehicle perspective.

According to a further expedient embodiment, during an automatic movement of the vehicle, learning data is determined from the first data and / or the second data, which data are transmitted to another vehicle for processing and / or transmitted to a central computer for automatic control of the object. In a spatially delimited area, algorithms which perform object-recognizing and / or measuring and / or classifying methods can also learn by matching with one another and / or by correcting with the second data of the outside sensor system. As a result, a vehicle manufactured, for example, in the spatially delimited area or moved for purposes of logistics can pass on its "experience" in the form of the evaluated first data to other vehicles. This learning data can be taken into account by a (in particular vehicle-external) computing unit or by other vehicles when performing automatic actions. Thus, both a vehicle and a controlled object in the spatially delimited movement may behave as if they already knew the environment. Deviations from an optimal process, which may be caused for example by unexpected objects or objects with another arrest, can be minimized step by step.

According to a further expedient embodiment, the first data determined by the vehicle sensor system and / or the second data determined by the external sensor system are transmitted to a computer unit for controlling a production device along or near which the vehicle moves. In the simplest case, a vehicle may stop or change in its operation via wireless communication with a stationary computer, a tape, or a device that may be dangerous to a person or property. For example, if the vehicle with its vehicle sensor system recognizes a person or an object that is to undertake a further processing step on the vehicle, certain preparatory steps can be automated by processing the control data.

According to a further expedient embodiment of the of the vehicle sensor system determined first data and / or the second data determined by the external sensor, a motion description and / or a motion recommendation and / or an interruption of a movement and / or an action instruction determined and transmitted to the one or more objects to be controlled. This embodiment of the method can be used to z. As a robot or other machining tool to show an efficient way to the vehicle, so that, for example, a processing step (eg, a rework) can be performed. The movement description and / or movement recommendation and / or action instruction may in particular have a spatial and / or temporal reference to the location on the vehicle at which a processing step is to be performed. The corresponding control data thus take into account where the vehicle or certain parts of the vehicle will be in the near future. The execution of a processing step thus becomes easier while the vehicle and possibly the object or objects move. The determination of the appropriately prepared control data can z. B. be transmitted from a central processing unit or a distributed system. The associated advantage is that with the aid of the vehicle sensor system and / or the exterior sensor system, the vehicle can carry out and if necessary monitor the control and execution of an action with respect to the vehicle.

According to a further expedient embodiment, a direction-dependent motion request is transmitted to the object or objects as control data from the first data determined by the vehicle sensor system and / or the second data determined by the external sensor system, wherein the motion request contains information about which Direction a respective object to be controlled to make a movement. The control data, in particular the motion request, may include the control of the objects in two dimensions. The control data can be the movement of the object in two mathematically orthogonal or independent directions of movement, z. B. "longitudinal" and "across" represent. Preferably, a motion request may include a description of trajectories for at least one object or two objects. The trajectories may be represented in the control data by a scientific context, e.g. Example by means of a chain of vectors, splines or formulas. The control data may also include quantitative measures of motion for objects, in particular velocity, acceleration, motion data for a rotation of the object to be controlled.

For example, in a possibly not yet finished motor vehicle built-in vehicle sensors is used to other events in the vicinity of the vehicle, eg. As a manufacturing environment to control. In particular, this may increase at a predetermined level increasing risk of collision with the vehicle, an automatic stopping and / or avoiding other objects, such. As forklifts, robots, other vehicles cause. In the simplest case, the vehicle can steer other objects that it detects with the help of its vehicle sensors with corresponding control data from its trajectory or tune trajectories with other objects. This allows a "choreography" of the vehicle and other objects. Preferred is a control with spatial reference, which is different depending on the object or object type.

The vehicle sensor system, which has been developed in series with great effort for a vehicle, is thus used to control potentially dangerous objects for the vehicle or people. Instead of emergency braking of the vehicle, as would be possible by a corresponding driver assistance system of the vehicle, the vehicle sensor system of the vehicle initiates an evasive or braking action which relates to the collision object. This can be more effective compared to controlling the vehicle. For example, in this way many more disturbances in the form of mutual blockings can be avoided or resolved in a time-efficient manner.

According to a further expedient embodiment, the data of one vehicle detected by the vehicle sensor system are compared with the data of another vehicle detected by the vehicle sensor system in order to determine whether these have an at least partially identical or similar movement trajectory with respect to a production device, wherein information about the movement trajectories is processed to obtain the control data. As a result, unplanned changes in the environment can be detected. Such changes can also be visualized by an arithmetic unit for a dispatcher. This makes it possible to detect changes in the environment, which avoids the risk of a collision between the vehicle and an object.

According to a further expedient embodiment, visual information from at least one vehicle camera, in particular a fused image, from the at least two vehicle cameras, for generating the control data, are processed as first data from the vehicle sensor system. This embodiment makes use of the all-round view of the vehicle, which is also known in part as a top view or a three-dimensional (3D) top view. Image data and / or object recognition data from a plurality of vehicles that are similarly ranked or by the same device, e.g. As a band to be moved, fused. For example, it is also possible to generate an "endless image" of a maneuvering area or of the belt, with the first data of the vehicle sensor system of a vehicle being added to one another at different times, in particular depending on the vehicle movement. In this case, the data from the vehicle sensor system and / or the first data can be assembled from partially overlapping spatial areas and / or from different vehicle positions. This can be configured in a way similar to the composition of a panorama photo.

Furthermore, it can be provided to update a once generated fused image with each further vehicle accordingly. Other vehicles or objects as well as a computer controlling the objects of the spatially delimited environment can thus use data of the vehicle or vehicles. A statistical result for the control of plant facilities and other objects can be provided.

In another embodiment, the frequency and / or the way in which the first data and / or the second data are detected by the vehicle sensor system or the external sensor system in dependence on an information or hypothesis on the position and / or orientation of the or adapted to the objects. For example, the sensing resolution and / or detection frequency in the environment of the vehicle may be increased. For this purpose, a Laser scanner after an initial clinging of the vehicle automatically shift the area with an increased resolution in the environment of the vehicle. As a result, in particular currently relevant object constellations with a higher resolution and / or from a suitable perspective, which offers as meaningful a representation of the object constellation, can be detected. At the same time there is an improved use of resources.

In a further embodiment, at least one sensor of the external sensor system can be moved at least partly automatically as a function of the vehicle movement and in particular corresponding to the vehicle movement. The outdoor sensor can be a stationary sensor. It preferably comprises a plurality of sensors, for. As cameras, which are arranged in a uniform or non-uniform grid, preferably at about the same height above the Rangierbereichs of the vehicle. It can, for. B. a laser scanner or a device for generating a structured light on a rail above a relevant area to be moved. Their movement is coordinated with the movement of the vehicle.

In a further refinement, the second data comprise partial data of a plurality of sensors of the external sensor located at different locations, with a weighting of the partial data being carried out depending on the position of the vehicle relative to these sensors. In this case, the weighting of the partial data of individual sensors as a function of the semi-automatic movement of the vehicle, in particular in small steps, can be changed. For this purpose, one or more spatial (transition) regions can be defined particularly expediently, in which or which the second data comprise partial data of specific sensors.

According to a further expedient refinement, partial data of two sensors of the external sensor system are fused if, during a movement of the vehicle, there is a transition of a detection of one of the two sensors to another of the two sensors. As a result, at least two sensors can mutually plausibility in a less well supplied with the data of a sensor area. Thus, the second data can also achieve the required data quality in a transition area.

In a further refinement, one or more probabilities of the presence and / or a specific one of the movements and / or a specific position of an object in the near future are determined for the determination of the control data. This makes it possible to consider non-secure information in the data fusion and thus also in the control of the object or objects. Thus, a relatively secure information can also be reconstructed from a plurality of insecure information or a relatively secure prognosis can be generated for the near future.

In a further embodiment, a prediction of the vehicle position and / or an object position for one or more times for the near future can be determined, wherein in the step of determining the control data one or more collision probabilities between different locations of the vehicle contour and the object contour are determined. In this way, different collision probabilities can be determined, which can occur with different movement options of the object or objects. Optimized control data can be determined, which overall gives a lower, optimized collision probability.

It is also possible to virtually simulate several variants of the movement of the object (s) and / or to determine the amount of risk for possible collisions between various contour points of the vehicle and of the object (s).

In a further refinement, spatial data which represent spatial conditions of the spatially delimited area can be processed in the determination of the control data. For this purpose, for example, a three-dimensional model of the spatially delimited area can be used. The common for many vehicles conditions, such. As physical space boundaries, position of the machines, security areas, etc., can be used as an existing data model (eg similar to CAD = Computer Aided Design). This allows the use of a previously accurately measured special environment. This results in a much higher security when processing information of the specific environment. This reduces the requirements with regard to sensor performance and costs.

As spatial data, information about the drivable and non-navigable spatial areas of the spatially delimited area, in particular in conjunction with temporal information about their respective validity, can be processed. A processor device of the vehicle and / or the spatially delimited area calculates z. Allowed or unauthorized movements using an existing spatial model of the specific environment. Possible routes can already be included and thus do not have to be detected by the object or objects. In the case of factory buildings or factory premises z. B. Hazardous areas with different levels of danger in the vicinity of robots or other Machines are represented in the data model. As a result, the process can be accelerated and realized with fewer resources. Also, a consideration of relatively abstract circumstances, such. B. of predefined and at certain times not passable security areas, take place, which can not be detected with the or the sensors.

As room data also information with spatial and temporal relation to already planned or possible movements of one or more objects can be processed at least within the possible movement space of the vehicle. This also makes it possible to relieve the strain on the vehicle sensor system and the external sensor system as well as the processing units performing the processing.

In a further refinement of the method, control data can be determined as a function of further data representing operating commands of a remote control device. This data can be provided by a computing unit of the vehicle or by a computing unit outside the vehicle. The further data representing operating commands of a remote control device are processed, in particular, in a control of the external sensor system and / or in planning in a vehicle trajectory and / or planning of an object trajectory and / or in determining the control data and / or in the correction of already determined control data , The further data can also be processed instead of the first and / or second data.

This makes it possible to realize an additional and / or a corrective intervention by a person who is outside the vehicle. The person can z. B. be a dispatcher and intervene in problem cases corrective in the automated process. The person can z. B. from a greater distance and additionally include information of the vehicle sensor and / or outdoor sensors in its decision to generate the other data. By processing the first and / or the second data for controlling the object or objects, however, interventions by a person can be kept low.

In this case, data representing operating commands of a remote control device can be taken into account as correcting the control data generated in dependence on the first data and second data. Thus, certain actions of a dispatcher, for. In relation to a control of the object or objects, by means of one or more additive and / or multiplicative and / or logarithmically correcting factors. Furthermore or additionally, the method may be configured such that one of at least two potentially possible movement variants of the object or objects is selected as a function of data representing operating commands of a control device.

Control data can also be determined as a function of the vehicle movement data, in particular odometric data. The vehicle movement data may be current or, in particular, predicted for a certain time interval or a time in the near future, vehicle movement data. Vehicle movement data can be detected or determined by means of the vehicle and / or by means of the external sensor system. In this case, the existing devices in the vehicle, for. As wheel sensors, steering sensors, etc., provide at least a portion of the data. The external sensor system or a local positioning system can provide at least part of the data representing odometric measurements of the vehicle. Preferably, the matching of the data, wherein the matching data in the implementation of the method are processed as data with a higher probability of statement than data that do not or only slightly match.

The invention also provides a computer program product that can be loaded directly into the internal memory of a digital computer and includes software code portions that perform the steps in accordance with the methods described above when the product is running on a computer. In connection with the method described above, a computer is to be understood as meaning one or more computing units of a vehicle and / or a computing or control unit of the external sensor system and / or one or more distributed computers for controlling components of the spatially delimited area.

Finally, a device is proposed for the automatic control of one or more objects within a spatially delimited area intended for the manufacture or maintenance or parking of one or more vehicles, wherein the vehicle has a vehicle sensor system with at least one sensor for detecting a part of the environment of the spatially delimited area, wherein the device is adapted to detect by means of the vehicle sensor system part of the environment of the vehicle within the spatially delimited area and to provide first data which the object or objects in the vicinity of the vehicle represent; Determine control data for controlling the movement behavior of the object or objects while processing the data provided by the sensor system; and a control of the or Make objects in dependence of the determined control data, which leads to a movement or to a change of the movement or a change of the state or behavior of the object or objects.

The spatially limited area for manufacturing or maintaining or parking a vehicle may be configured by the method such that vehicles and / or objects move in different directions, especially outside limited fixed directions of movement (also independent of belts, guide rails such as in a conventional factory or a car wash). The capacity of the devices for the manufacture, care of vehicles or serve an available area can be used much better and more flexible

The device has the same advantages as described above in connection with the method according to the invention.

Furthermore, the device may comprise further means for carrying out the method described above.

The invention will be explained in more detail below with reference to an embodiment. Show it:

1 a schematic representation of a device according to the invention for the automatic control of one or more objects, and

2 a schematic representation of the sequence of the method according to the invention.

1 shows a schematic representation of a device according to the invention for the automatic control of one or more objects within a spatially delimited area 1 , The spatially delimited area is, for example, a spatially closed and / or thought-out area, such. B. a production hall for vehicles. Alternatively, the spatially delimited area may be a parking garage or a parking area and the like. For example, the spatially delineated area could also be a parking garage or a carwash.

In the spatially delimited area 1 For example, only two vehicles move 10 . 20 along a common direction of movement 50 , The vehicles 10 . 20 For example, by a not-shown band, as this is used for the production of vehicles, in the direction of movement 50 to be moved.

Each of the vehicles 10 . 20 includes a respective vehicle sensor system 11 . 21 , The vehicle sensor systems not shown in detail 11 . 21 comprise a number of sensors which serve to detect the surroundings of the respective vehicle. The environment detection can serve a sensing, surveying or detection of objects outside the vehicle. Such vehicle sensors are usually part of driver assistance systems. The sensors used in each case can comprise in a known manner one or more cameras, radar, ultrasound sensors, a stereo camera or a light-emitting sensor system. The latter can for example be designed as a laser scanner.

The processing of the respective vehicle sensors 11 . 21 recorded data is typically done by a not-shown processing unit of the vehicle in question 10 . 20 , In the context of the method described in more detail below, the processing of the respective vehicle sensor systems 11 . 21 provided by one or more computing units outside the vehicle or vehicles 10 . 20 respectively.

In the vicinity of the two exemplified vehicles 10 . 20 are exemplary three objects 41 . 42 . 43 shown. The objects may each be a robot, a forklift, a transport and loading device for motor vehicles and the like. Generally, an object could 41 . 42 . 43 also by other motor vehicles or persons in the vicinity of the vehicle or vehicles 10 . 20 stop, be represented. In the following description it is assumed that the objects 41 . 42 . 43 processing of corresponding control data with regard to their movement (ie direction of movement and / or movement speed) or a change of their state or behavior (eg switched on or configured, in a first state or a second state, a state being, for example, a certain one Position of a robot arm and the like, etc.) are controllable. A control of respective objects 41 . 42 . 43 can be done, for example, in terms of their longitudinal dynamics, lateral dynamics, vertical dynamics, stabilization or rolling motion. A control can for example be such that a sequence of predetermined movements or maneuver elements takes place, which, for example, a simultaneous and / or temporally successive change a longitudinal dynamics and lateral dynamics by itself. The objects 41 . 42 . 43 have for processing this control data optionally a corresponding arithmetic unit, which may be an integral part of the respective object. Alternatively, the processing of the control data also by one outside the objects 41 . 42 . 43 arranged, not shown processing unit done.

In the spatially delimited area 1 is also an outdoor sensor 30 intended. This includes, by way of example only, a number of six individual sensors or sensor systems 31 , ..., 36 , The sensors or sensor systems 31 , ..., 36 can be stationary or movable in the spatially delimited area 1 be arranged. The respective sensors or sensor systems 31 , ..., 36 are designed to, the or the vehicles 10 . 20 as well as the environment of the vehicles from outside the vehicles 10 . 20 capture. These are schematically respective detection areas 31a , ..., 36a for each of the sensors or sensor systems 31 , ..., 36 shown. The of the respective sensors or sensor systems 31 , ..., 36 recorded second data are processed by one or more, not shown processing units. This arithmetic unit is designed to generate a data model of the spatial reference environment. Sensor types are the sensors or sensor systems 31 , ..., 36 preferably light-emitting sensors, such as. B. laser scanner. The sensors or sensor systems 31 , ..., 36 are preferably perceptive sensors. It is basically not necessary that all of the sensors or sensor systems 31 , ..., 36 the outdoor sensor 30 are of the same type. Rather, different sensor principles can be used.

The above-mentioned processing units for processing the first data and the second data and for processing the control data for controlling the objects 41 . 42 . 43 may be separate arithmetic units that are communicatively connected to each other. However, the arithmetic units can also be a common arithmetic unit or a distributed computer system.

With reference to 2 the method according to the invention is described in more detail below. In a first step S1, a detection of a first object constellation takes place with the aid of the respective vehicle sensor systems 11 . 21 (Which are preferably perceptive vehicle sensor systems) from the respective vehicle perspective of the vehicle 10 . 20 , The from the detection of vehicle sensor systems 11 . 21 resulting data is provided as first data. Through the vehicle sensor systems 11 . 21 Objects that are in the direction of movement of the vehicle can be detected very well. For example, detects the vehicle sensors 11 of the vehicle 10 the objects 41 . 42 , The object 43 is in the illustrated embodiment by the vehicle sensors 21 of the vehicle 20 detected.

Preferably, at the same time, in steps S2 and S3, a detection of a second object constellation takes place with the aid of the external sensor system 30 , A part number of sensors 31 , ..., 36 detected in step S2, the second object constellation from a first outside perspective. A second part number of sensors 31 , ..., 36 detected in step S3, the second object constellation from a second external perspective. Depending on the sensors or sensor types used, objects can be detected that are obscured by other objects. The exterior sensor 30 In addition, it allows for the detection of objects that are in the ultra-near range of each of the vehicles 10 . 20 are located. For example, ground marks (in 1 not shown). The from the detection of outdoor sensors 30 resulting data is provided as second data.

The objects of the first object constellation and the objects of the second object constellation are the same objects which are detected by the respective sensor systems from different perspectives.

Appropriately, the detection of the first and the second object constellation with the respective vehicle sensor system takes place 11 . 21 or the respective outdoor sensor 30 synchronized in time. The synchronization can z. B. be ensured by means of a synchronization signal which (preferably wirelessly) from the already described computing unit to the respective vehicle 10 . 20 and the outdoor sensors 30 is transmitted. The first and second data can be properly allocated to each other and at specific times by the timing synchronization. As described, an object constellation can also be detected, which includes the vehicle itself and certain ground markings, such as are frequently present in the environment of production facilities. For example, the distance of a vehicle culture to a ground mark can thereby be detected.

In an optional, in 2 not shown, a step of providing coordinates of vehicle parts of the vehicle or vehicles 10 . 20 and / or one or more of the objects 41 . 42 . 43 done by a local tracking system. An orientation of the vehicles is preferred 10 . 20 detected as an angle value with respect to a given movement axis or a fixed axis of the vehicle or object. The detection and provision of such coordinates (in general: position detection) preferably takes place synchronously with the detection of the second data by the vehicle sensor systems 11 . 21 and the outdoor sensors 30 , This in 1 not shown local location system is too a position determination / detection of the vehicle or vehicles 10 . 20 and / or one or more of the objects 41 . 42 . 43 in the spatially delimited area 1 educated. This provides coordinates in an internal coordinate system valid for this area. The local location system enables a higher accuracy and reliability of the position determination in such spatial parts of the spatially delimited area 1 in which the vehicle sensors 11 . 21 or the external sensor system 30 can not provide sufficient data quality. The determination of control data described below for controlling the movement and / or behavior of one or more of the objects 41 . 42 . 43 can also be done depending on the provided coordinates.

The first and / or the second data expediently comprise information about the perspective from which the data was acquired. Optionally, the first and / or the second data also comprise the information about the spatial regions which are represented in the first and / or in the second data. This assignment of the information about the perspective to the acquired data can be made on the basis of the type of sensor, the orientation and its detection range.

In step S4, a match between at least two objects is determined by a data fusion of the first and the second data. The data fusion provides third data as the result. Here, the information from the first and the second data is also assigned to specific objects. In particular, in the step of data fusion, a conversion to a standardized perspective can take place. This z. B. Perspectives on an object 41 . 42 . 43 or vehicle 10 . 20 to a certain perspective, eg. B. a vertical supervision, to be converted. This standardization can simplify the identification of the data fusion.

Furthermore, the data of the local location system and the first data and / or the second data and / or the fused third data can optionally be assigned to an object constellation as a function of their reference to concrete objects. This process can be done iteratively in several steps by adding to each object 41 . 42 . 43 corresponding features in the first and / or second and / or third data are determined.

Based on the merged third data, at least one object constellation is determined in step S5. Preferably, such an object constellation is determined, which for a current or future movement of the object or objects 41 . 42 . 43 is relevant. Relevant are such objects or object constellations, with which a respective vehicle 10 . 20 will interact with a certain probability. An interaction can not only an undesirable interaction, such. As a collision, but also be a desired interaction. In particular, the interaction may be a physical interaction, e.g. B. also an automated loading of the respective vehicles 10 . 20 , A possible interaction between a particular vehicle 10 . 20 and one of the objects 41 . 42 . 43 is determined in step S6 by the determination of probability values.

In step S7, the control data is determined to control at least one movement and / or the behavior of one or more of the objects 41 . 42 . 43 depending on the fused, third data, which - as described - indirectly dependent on the first data and the second data. For this purpose, in step S7 first a respective movement trajectory for each of the objects and / or the vehicles 10 . 20 planned and determined The control data are determined in step S8 under processing of the vehicle trajectory such that a collision risk between the objects and the vehicles is minimized and one or more predetermined, desired interactions are achieved. Subsequently, in step S8, the activation of one or more actuators of the object or objects takes place 41 . 42 . 43 , and optionally also the one or more vehicles 10 . 20 depending on the determined control data.

As part of the determination of the control data and / or in the control of at least one actuator of the object or objects 41 . 42 . 43 , and optionally one or more of the vehicles 10 . 20 the data of the local positioning system can be taken into account. This can increase the precision and safety of the partially automated movements.

Performing a desired interaction with an object can be done in one or more steps. The interaction is detected in particular by the vehicle sensor system and / or the external sensor system. In particular, the progress of the interaction as an object constellation is recorded at continuous times. Thus, a stepwise change of the corresponding object constellation during the interaction can be detected and result in changed control data.

The control data is determined or corrected in dependence on the determined first data and the second data representing the object constellation such that an improved interaction with the object or objects 41 . 42 . 43 in interaction with the movement of the vehicle or vehicles 10 . 20 is reached. At the same time the control data are determined such that an undesirable interaction with an object, such. As a collision that could lead to damage to the vehicle and / or one of the objects or mutual blockages, in which at least one object and / or vehicle is at least temporarily prevented in its further movement, be avoided.

Is one or more of the sensors 31 , ..., 36 variable or mobile in terms of its coverage, these may vary depending on the movement of the vehicle or vehicles 10 . 20 and / or depending on the movement of one or more of the objects 41 . 42 . 43 to be controlled. For example, a panning camera can change its position or orientation. This may, for example, also be a function of a change in the control data which leads to a change in the movement and / or behavior of one or more objects 41 . 42 . 43 leads done. The control of outdoor sensors 30 depending on the control data is advantageous because the external sensors already for detecting future positions of the vehicle or vehicles 10 . 20 and / or the objects 41 . 42 . 43 is controlled. This will cause the movements of the outdoor sensors 30 as well as the moving objects in the spatially delimited area 1 synchronized so that the outside sensory 30 already be aligned to a future, planned entry position. Such a procedure facilitates data processing and thus speeds up the process.

For repetitive movements, ie predetermined movements of the vehicles 10 . 20 and / or one or more of the objects 41 . 42 . 43 can be obtained from the determined data learning data by information on features of an object constellation, which were represented in the data already obtained or were determined, information is assigned to what extent this representation of the object constellation is correct or by a result of another step of Have found the procedure to be correct or precise, or incorrect or imprecise. Determined learning data can be stored in a database, not shown in the figures. In the further course of the method, the learning data can be read from the database. Depending on this data, the external sensor system and / or the vehicle sensor system can be controlled and the corresponding data can be determined and evaluated. This optimizes a procedure of the method on the basis of predefined criteria.

LIST OF REFERENCE NUMBERS

1

spatially delimited area

10

first vehicle

11

Vehicle Sensors

20

second vehicle

21

Vehicle Sensors

30

External sensors

31

first sensor of outdoor sensors 30

31a

Detection range of the first sensor 31

32

second sensor of the external sensor 30

32a

Detection range of the first sensor 32

33

third sensor of the outdoor sensor 30

33a

Detection range of the first sensor 33

34

fourth sensor of the outdoor sensor 30

34a

Detection range of the first sensor 34

35

fifth sensor of the outdoor sensor 30

35a

Detection range of the first sensor 35

36

sixth sensor of outdoor sensors 30

36a

Detection range of the first sensor 36

41

first object

42

second object

43

third object

50

Direction of movement of the first and second vehicle

S1

step

S2

step

S3

step

S4

step

S5

step

S6

step

S7

step

S8

step

Claims (20)

Method for automatically controlling one or more objects ( 41 . 42 . 43 ) within a spatially delimited area ( 1 ) used for the manufacture or maintenance or storage of a vehicle ( 10 . 20 ) is provided or formed, wherein the vehicle ( 10 . 20 ) a vehicle sensor system ( 11 . 21 ) with at least one sensor ( 31a . 32a . 33a . 34a . 35a . 36a ) for detecting a part of the environment of the spatially delimited area ( 1 ), in which - the vehicle sensor system ( 11 . 21 ) a part of the environment ( 31a . 32a . 33a . 34a . 35a . 36a ) of the vehicle ( 10 . 20 ) within the spatially delineated area ( 1 ) and provides first data which the object or objects ( 41 . 42 . 43 ) in the vicinity of the vehicle ( 10 . 20 represent); Control data for controlling the movement behavior of the object or objects ( 41 . 42 . 43 ) under processing of the vehicle sensors ( 11 . 21 ) are provided; and - a control of the object or objects ( 41 . 42 . 43 ) takes place as a function of the determined control data which leads to a movement or to a change in the movement or a change in the State or behavior of the object or objects ( 41 . 42 . 43 ) leads. Method according to Claim 1, in which the spatially delimited area ( 1 ) an external sensor ( 30 ) with at least one sensor for detecting part of the environment ( 31 . 32 . 33 . 34 . 35 . 36 ) of the spatially delineated area ( 1 ), wherein the external sensors ( 30 ) provides second data representing the vehicle ( 10 , 20 ) and / or one or more objects ( 41 . 42 . 43 ) in the environment of the vehicle. Method according to Claim 1 or 2, in which the activation of the object or objects ( 41 . 42 . 43 ) for achieving a predetermined interaction with one or more of the objects ( 41 . 42 . 43 ) is carried out. Method according to Claim 2 or 3, in which the first data comprises at least one object constellation of in each case at least two objects ( 41 . 42 . 43 ) and the second data comprises an object constellation from the vehicle ( 10 . 20 ) and at least one of the objects ( 41 . 42 . 43 ). Method according to one of the preceding claims, in which the detection of the surroundings of the vehicle by the vehicle sensor system ( 11 . 21 ) and by the external sensors ( 30 ) takes place in a temporal dependency, in particular synchronized. Method according to one of the preceding claims, in which the control data are determined as a function of the first data and the second data, wherein third data are determined and provided from a comparison of the first data and the second data which relate to the same object ( 41 . 42 . 43 ) or the same object constellation. Method according to Claim 6, in which, when the first data and the second data are compared, the respective detection perspective of an object ( 41 . 42 . 43 ) by the external sensors ( 30 ) and / or by the vehicle sensor system ( 11 . 21 ) is taken into account. Method according to one of the preceding claims, in which by comparing with the respective vehicle sensor systems ( 11 . 21 ) of several vehicles ( 10 . 20 ) determined first control data are determined by the at least partially automatic change of the control of the object or objects ( 41 . 42 . 43 ) is achievable by generating learning data. Method according to one of Claims 3 to 9, in which the external sensor system ( 30 ) above and / or below a part of the area ( 31a . 32a . 33a . 34a . 35a . 36a ) in which the vehicle ( 10 . 20 ) is arranged, so that the vehicle ( 10 . 20 ) is sensory detected from a top view and / or bottom view of straight or obliquely above or below. Method according to one of the preceding claims, in which at least part of the determination of the control data, in particular the determination of a trajectory of the object or objects ( 41 . 42 . 43 ) and / or not by the object or objects ( 41 . 42 . 43 ) passable spatial areas of the spatially delimited manufacturing area ( 1 ), in a processor unit outside the vehicle ( 10 . 20 ) he follows. Method according to one of the preceding claims, in which during an automatic movement of the vehicle ( 10 . 20 ) are determined from the first data and / or the second data learning data, which to another vehicle ( 10 . 20 ) for processing and / or for the automatic control of the object ( 41 . 42 . 43 ) are transmitted to a central computer. Method according to one of the preceding claims, in which the vehicle sensors ( 11 . 21 ) determined first data and / or by the external sensors ( 30 ) determined second data to a computing unit for controlling a manufacturing device, along or near the vehicle ( 10 . 20 ) are moved. Method according to one of the preceding claims, in which, as control data from the vehicle sensors ( 11 . 21 ) determined first data and / or by the external sensor ( 30 ), a movement description and / or a movement recommendation and / or an interruption of a movement and / or an action instruction are determined and sent to the object (s) to be controlled ( 41 . 42 . 43 ) be transmitted. Method according to one of the preceding claims, in which, as control data from the vehicle sensors ( 11 . 21 ) determined first data and / or by the external sensor ( 30 ), a direction-dependent motion request to the object or objects ( 41 . 42 . 43 ), wherein the motion request contains information about the direction in which a respective object to be controlled ( 41 . 42 . 43 ) should make a move. Method according to one of the preceding claims, in which the vehicle sensors ( 11 . 21 ) recorded first data of a vehicle ( 10 . 20 ) with those of the vehicle sensors ( 11 . 21 ) recorded first data of another vehicle ( 10 . 20 ) to determine if they are in With respect to a manufacturing device having an at least partially the same or similar movement trajectory, wherein information about the Bewegungsstrajektorien is processed to determine the control data. Method according to one of the preceding claims, in which first data from the vehicle sensor system ( 11 . 21 ) a visual information of at least one vehicle camera, in particular a fused image from at least two vehicle cameras, is processed to generate the control data. Method according to one of the preceding claims, in which first data from the vehicle sensor system ( 11 . 21 ) Data of a radar system and / or an ultrasound system and / or a light-emitting system are processed to generate the control data. A computer program product that can be loaded directly into the internal memory of a digital computer and includes software code portions that perform the steps of any one of the preceding claims when the product is run on the computer. Device for automatically controlling one or more objects ( 41 . 42 . 43 ) within a spatially delimited area ( 1 ) used for the manufacture or maintenance or storage of a vehicle ( 10 . 20 ) is provided / formed, wherein the vehicle ( 10 , 20 ) a vehicle sensor system ( 11 . 21 ) with at least one sensor ( 31a . 32a . 33a . 34a . 35a . 36a ) for detecting a part of the environment of the spatially delimited area ( 1 ), wherein the device is designed to - by means of the vehicle sensor system ( 11 . 21 ) a part of the environment of the vehicle ( 10 , 20 ) within the spatially delineated area ( 1 ) and provide first data which the object or objects ( 41 . 42 . 43 ) in the vicinity of the vehicle ( 10 . 20 represent); Control data for controlling the movement behavior of the object or objects ( 41 . 42 . 43 ) under processing of the sensors ( 11 . 21 . 30 ) to provide data provided; and - a control of the object or objects ( 41 . 42 . 43 ) in response to the determined control data which leads to a movement or to a change of the movement or a change of the state or behavior of the object or objects ( 41 . 42 . 43 ) leads. Apparatus according to claim 19, which comprises further means for carrying out the method according to any one of claims 2 to 18.

Images