This disclosure relates to a self-teachable robotic station for handling and processing objects such as semiconductor wafers, disks, substrates, and other small and delicate objects which are not necessarily round or flat, e.g., rings, etc. More specifically, the disclosure relates to teachable robotic stations of the aforementioned type provided with a function of self-compensation of errors accumulated as a result of slacks, wear, flexibility in connections, etc.
Robotic stations with self-teaching functions are known in the art. One of the most time-consuming and difficult tasks in connection with the use of robotic stations is a robot arm with End of Arm Tools (EOAT), such as edge grippers, or end effectors for moving along a preprogrammed rout with high accuracy on each working cycle.
A large automated robotic station may have a plurality of points that must be manually taught. The operator moves a robot's end effector through a required number of degrees of freedom to align the end effector within an acceptable tolerance to given picking/placing/processing positions. The speed and accuracy of this operation depends on such factors as experience, fatigue, visual acuity of the robot operator, etc. Even though the teaching process is accurate, reliability of the robot operation may also depend on plays in connections of the tooling and end effectors to the robot arm and on slack accumulated after a certain amount of the performed cycles. Consequently, such taught points often need to be refined by the robot operator one or more times to increase the accuracy of the point.
The precision teaching allows accurate placement of delicate parts into cassette or other media, or on process chuck, etc., without damaging and/or rubbing walls thus reducing generation of particles and improving the yield.
This invention makes possible flexible adaptation of the principle of Smart Factories to industrial production in the Industry 4.0 environment. The Industry 4.0 is the principle of application of Internet of Things and services to manufacturing at the fourth stage of industrialization which is currently revolutionizing the manufacturing engineering sector. Hence, cyber-physical systems (CPS) improve resource productivity and efficiency and enable more flexible models of work organization. In Industry 4.0, robots and humans will work hand in hand and in these environment teachable robotic stations will play a significant role.
Teachable robotic stations for handling and processing objects are known in the art and described in a number of patent publications.
For example, U.S. Pat. No. 5,297,238 discloses a method for calibrating a tool control frame (TCF) on a robot with respect to a known calibration reference claim (CRF), wherein the (CRF) is in rigid body relationship with a robot link. The method includes the steps of (a) attaching a sensory tool to the robot link, (b) calibrating the sensory tool with appropriate units of measurement, (c) identifying a calibration feature to be mapped by the sensory tool, (d) causing relative movement of the calibration feature within sensing range of the sensory tool, (e) recording robot configuration and pose as a first data record, (f) causing relative movement of the calibration feature and sensory tool along a known direction to a new position within sensing range, (g) recording configuration and pose as a second data record, and (h) applying coordinate transformations to the respective data records to generate a correct tool control frame (TCF) pose with respect to the (CRF).
U.S. Pat. No. 4,675,502 discloses a real time tracking control for taught path robots. A real time steering capability is provided to permit robot motion to be modified continuously in three dimensions as the robot is moving along a taught path. An arc welding robot or other taught path robot has a sensor located on the robot arm to sense the position of a desired path. The tracking control provides real time steering commands to the standard robot taught path and are calculated based on maintaining a constant, preprogrammed velocity along the desired path and coordination with the taught path.
U.S. Pat. No. 4,831,549 discloses a device and method for improving orientation and/or location accuracy of a programmable robot with respect to a target object. The method consists of calibrating the position of a terminal control frame associated with a robot end-effector which is coupled to a robot distal link. Separated reference positions external from the robot are identified, as to geometry and spatial data. This identification data is stored for later recall and comparison for use in determining a localized relative frame of reference. The robot end-effector is moved to a first reference position and a rigid body error correction is determined. This correction is stored in computer memory for application to later computer movement.
International Patent Application Publication WO 2010136961 discloses a control device for controlling a robot having a robot arm with a number of individual arm sections, an end effector connected to one of the arm sections and a number of actuators for moving at least the end effector and at least one of the arm sections in at least two different modes of operation, i.e., a working mode and a training mode that corresponds to the working movements. The robot control device corrects the teaching position of the motion program stored in the storage, based on a change in the relative position obtained by the position calculator.
U.S. Pat. No. 8,242,730 discloses an automated robot teach tool and a method of use of the teach tool. The latter enables automatic teaching of pick and place positions for a robot. The automated robot teaches tool obviates the need for manual operation of the robot during the teaching. The result is an automated process that is much faster, more accurate, more repeatable and less taxing on a robot operator. The teach tool comprises: a body assembly with a proximity sensor mounted therein that is releasably mated with a robot end effector; a foot assembly coupled to the body assembly, wherein the foot assembly comprises a sensor target mounted therein that is located about the proximity sensor; and wherein the proximity sensor and the sensor target are configured to detect signals representative of a perturbation as the robot end effector moves the body assembly and foot assembly from a central position within a workpiece receptacle through six degrees of freedom. The aforementioned signals are used to determine a precise orientation for the robot end effector to pick up and place a workpiece to and from the workpiece receptacle. Signals representative of the perturbation are generated in response to the foot assembly colliding with a horizontal surface or a vertical surface of the workpiece receptacle.
The present invention relates to a teachable robotic station for handling and processing objects such as semiconductor wafers, disks, and substrates. The objects are not necessarily flat and may comprise small and delicate objects other than disks, or flat substrates. The robotic station of the present application consists of a frame with a platform that supports an industrial robot arm having a plurality of axes for linear and rotational motions, e.g., a 4, or 5, or 6-axis robot arm, which allows attachment of interchangeable working tools such as end effectors with grippers, vacuum handlers, measurement tools, or other changeable components. In contrast to conventional robotic stations of the same class, the robotic station of the invention is characterized by greater versatility and ability of processing several objects of different kinds in one working cycle. For example, such objects may comprise semiconductor disks and interleafs such as paper, plastic, or fabric interleaves, separation rings; bridge tools for processing wafers having different sizes, outside diameters and/or thicknesses, etc. For this purpose, an industrial robot arm may be coupled with various End of Arm Tools (EOATs) such as changeable end effectors, or the like. When one of these EOATs performs an operation, others EOATs may be maintained in a waiting state with or without an object to be treated.
Another distinctive feature of the robotic station of the invention is a provision of a set of changeable tools that can be replaced or interchanged within the same working cycle automatically for performing appropriate operations in an appropriate sequence.
According to another aspect of the invention, the robotic station is provided with a unique self-teaching system that allows precision teaching of the robot arm and all machine stations for each EOAT where the object(s) can be placed and picked to perform multiple working cycles in accordance with a required sequence, with high accuracy and automatic compensation of an accumulated error and without interruption of the working process. Such an error may be accumulated due to stack of tolerances and object holding media deviations after completion of a certain number of working cycles or because of a play in attachment mechanisms for connection of changeable tools, etc.
Teachable robotic stations for handling, transporting, assembling, bonding and processing various objects are known in the art. In the inventors' opinion, the robotic station of the invention differs from the known stations of this type in that the robot teaching system is based not on measurement of the objects to be treated but on memorizing specific positions which the robot's tools and the objects assume for performing required operations. For this purpose, two reference objects are used, i.e., a stationary flexible reference tactile (touch) sensor for initiation of the self-teaching operation and a second stationary reference object, e.g., a spherical precision ball rigidly secured on the platform of the robotic station in the vicinity of the fixed-position flexible reference tactile sensor. The tactile (touch) sensors are well known devices that are used in coordinate measuring machine (CMM), CNC machine tools, etc. An example of such a sensor is a modified touch probe of model TP20 of MSC, N.Y., Renishaw brand.
As any teachable robotic station, the station of the invention is provided with a central processing unit (CPU). The station also has a changeable tactile sensor that has the same coupling construction as any tool of the interchangeable EOATs and can be connected to the robot arm in the same manner as any tool. In fact, the tactile sensor is one of changeable working tools that is stored in a certain position on the platform and is accessible for automatic connection to and disconnection from the robot arm. In a predetermined period of time or after a predetermined number of working cycles the robot arm may change a current working tool with the tactile sensor for checking the state of accuracy in positions of all units and tools involved in the process. For this purpose, the end of the robot arm (herein referred to as a “coupler”) touches the stationary (non-changeable) flexible reference tactile sensor as many times as need for determining exact position of the coupler relative to the position of the changeable tactile sensor in a fixed, i.e., an auxiliary coordinate system of the platform. This coordinate system is introduced at the design and manufacturing stages of the robotic station, and coordinates of all components of the robotic station constantly fixed on the platform are known and stored in the CPU memory. In this auxiliary coordinate system, a position of the center of the stationary (non-changeable) flexible reference tactile sensor is assumed as a center of coordinate.
Next, the coupler goes to the changeable tactile sensor which has a shank insertable into the central opening of the coupler where the shank of the changeable tactile sensor is fixed by a locking mechanism, e.g., by spring-loaded balls. Next, the robot arm with the changeable tactile sensor fixed in the coupling touches the second reference object, i.e., precision ball, at least in four different points which allow to define a position of the ball center which is then assumed as a center of an operational coordinate system. In order to define a horizontal plane the probe touches 3 points of a tooling plate, and the plate is leveled in horizontal plane. Following this, the changeable tactile sensor touches the units and tools that participate in the specific process in so many points as necessary for unequivocally defining the positions of these parts and units in the aforementioned operational coordinate system.
In case of switching to commissioning of media of different manufacturers as well as at initiation of a new working cycle of the robotic station, positions of the stations, canisters, cassettes, etc. should be re-taught with the use of tactile sensors and reference objects.
After expiration of the given period of time or completion of a predetermined number of working cycles, or after service of robot, or power outage, the current working tool is placed in its proper place on the platform, the coupler of the robot arm is disconnected from the current working tool and is moved to the stationary (non-changeable) flexible reference tactile sensor to touch the probe. After determining the position of the stationary (non-changeable) flexible reference tactile sensor in the auxiliary coordinate system of the platform the self-teaching procedure of the robotic station is resumed according to the same scenario as described above.
All the operations are performed under control of CPU and periodically repeated. Thus, the periodically accumulated errors are eliminated and the self-teaching is automatically repeated in accordance with a given sequence.
This invention relates to a teachable robotic station for handling and processing flat or non-flat objects such as semiconductor wafers, disks, substrates, rings, bridge tools, etc. for processing wafers having different sizes, etc.
A general perspective view of a robotic station 20 of the invention is shown in
As can be seen from
All manipulations of the industrial robot arm 38 are possible due to degrees of freedom, six in the illustrated case, provided by the mechanism of the robot assembly 26 that may comprise a standard robot arm assembly engageable with the interchangeable end effectors 34 and 36 and other interchangeable EOATs, including sensors as teaching tools, that can be connected to and disconnected from the robot arm 38.
The robot end can interchangeably interact either with a robot arm head 40 which in
In fact, both the end effectors 34 and 36 and the robot arm head 40 are EOATs or working tools of the robot assembly which perform different functions. For example, the end effectors 34 and 36 are used for manipulating rigid flat objects such as semiconductor wafers or wafer substrates, and the robot arm head 40 is used for handling soft objects such as interleaves and/or rigid objects such as wafers or rings.
Reference numeral 42 designates a kinematic mount waiting station for replaced robot arm head 40 which stays on the station 42 while the robot arm 38 works with the end effectors 34a and 34b. Reference numeral 44 designates an intermediate storage for flat objects, e.g., wafers (only one wafer W is shown in
The robotic station 20 is also provided with an optical sensor 46 (
The robotic station 20 is also provided with a visual display 48 (
Let us consider operations of the robotic station 20 of the invention by referring to picking up, transporting, processing, and sorting semiconductor wafers W stored in the canisters 31 and 33 (
The working cycle is started from detecting a type of the objects located, e.g., in the canister 31, by an optical sensor 46 (
Upon completion of a given quantity of full working cycles or upon expiration of a given operational time, the robotic station 20 is switched to a self-teaching procedure.
More specifically, according to another aspect of the invention, the robotic station is provided with a unique robot teaching system that allows precision self-teaching of the robot arm and all machine stations for each EOAT where the object(s) can be placed into their storage locations or working positions and picked up to perform multiple working cycles in accordance with a required sequence, with high accuracy and with automatic compensation of an accumulated error without interruption of the working process. Such an error may be accumulated due to stack of tolerances and object holding media deviations after completion of a certain number of working cycles or because of a play in attachment mechanisms for connection of changeable tools, etc.
Having described the structure of the robotic station 20 of the invention, let us consider now the self-teaching system used in conjunction with the robotic station 20 and operation of its tools.
Teachable robotic stations for handling, transporting, and processing various objects are known in the art. In the inventors' opinion, the robotic station of the invention differs from the known stations of this type in that the robot self-teaching system is based not on measurement of the objects to be treated but on memorizing specific positions which the robot tools and the objects should assume for performing required operations. Another distinction is the use of a stationary tactile sensor, a changeable tactile sensor, and a fixed reference object which allow at each new self-teaching cycle to assign and form a new operational coordinate system. The measurement units used in this coordinate system are defined in terms of dimensions of the reference object.
More specifically, two reference objects are used, i.e., a first reference object in the form of (non-changeable) flexible reference tactile sensor 50 for initiation of the self-teaching operation (active sensor) and second reference object, e.g., a second stationary reference object such as a precision ball 52 rigidly secured on the platform 24 (only a part of this platform is shown in
The platform 24 comprise a monolithic plate, while reference teaching components, i.e., the reference tactile sensor 50 and the precision ball 52 may be installed on a changeable sub-platform 24a which is shown in
As any teachable robotic station, the station of the invention is provided with a central processing unit (CPU) (
In more detail, positions of the first and second reference objects, i.e., the stationary (non-changeable) flexible reference tactile sensor 50 and the second reference object, i.e., a precision ball 52 secured on the platform 24, are shown in
In a predetermined period of time or after a predetermined number of working cycles the robot arm 38 may replace a current working tool, e.g., the end effector 36, with the changeable tactile sensor 54 for checking the state of accuracy in positions of all units and tools involved in the process. For this purpose, the coupler 38a touches the stationary (non-changeable) flexible reference tactile sensor 50 as many times as needed for geometrically determining exact position of the coupler 38a relative to the position of the changeable tactile sensor 54 in an auxiliary fixed coordinate system of the platform 24, which was preliminarily stored in the memory of the CPU. In this auxiliary coordinate system, the position of the center of the stationary (non-changeable) flexible reference tactile sensor 50 is assumed as a center of coordinate, and initial coordinates of the coupler 38a and of the changeable tactile sensor 54 stored in its nest 34a of the platform 24 are known. More specifically, the coordinates of all units and stations fixed relative to the platform 24 are preset in the memory of the CPU. These data are inputted in the form of coordinates of the auxiliary coordinate system to the CPU directly from the production drawings or a reference model.
Next, the coupler 38a goes to the changeable tactile sensor 54, which has a shank insertable into the central opening of the coupler where the shank of the changeable tactile sensor is fixed by a locking mechanism 38b in a known manner, e.g., by spring-loaded balls 38c (
Next, the robot arm 38 with the changeable tactile sensor 54 fixed in the coupling 38a touches the second reference object, i.e., the precision ball 52, at least in four different points which allow to determine a position of the ball center which is then assumed as a center of an operational coordinate system. Following this, the changeable tactile sensor 54 sequentially touches all units and tools that participate in the specific process in so many points as necessary for unequivocally defining positions of these parts and units in the aforementioned operational coordinate system. This operation is illustrated in
After expiration of the given period of time or completion of a predetermined number of working cycles, the current working tool is dropped in its proper nest on the platform 24; the coupler 38a of the robot arm 38 is disconnected from the current working tool and is moved to the stationary (non-changeable) flexible reference tactile sensor 50 to touch the latter (
All the performed operations are displayed on the display 48 (
Thus, the modes of operation of the robotic station 20 can be described by the following four configurations. The first configuration relates to a condition when the coupler 38a of the robot arm 26 is free of any EOATs. This state corresponds to the condition directly before the start of the self-teaching procedure or directly after reset of data after elimination of the accumulated errors. The second configuration is a state at which the coupler 38a of the robot arm 38 is equipped with an appropriate EOAT such a moveable tactile sensor 54. The third configuration is a state at which the moveable tactile sensor 54 is replaced with a robot arm head 40, and the fourth configuration corresponds to a state when the couple 38a is equipped with a working tool such as, e.g., an end effector 36. Four configurations are given only as an example, as well as the number of EOATs used in the process may be greater or smaller than in the illustrated example. In other words, the robotic station 20 may have five, six, or more configurations but no less than two.
Although the invention was described with reference to specific examples of the robotic station components, it is understood that any changes and modifications are possible without departure from the scope of the attached patent claims. For example, the working tools and processing stations are not limited to those described and shown in the drawings. The stations and tools may comprise metrological device for weighing, measuring warping, flatness, bowing, etc. The stations and tools may be used not only unpacking, measuring and sorting but also for packing. The teachable robotic station of the invention is applicable for processing articles other than flat objects and can be use for processing bridges, small and delicate three-dimensional objects, small spherical objects, etc. The precision self-teaching robotics systems can be implemented in assembly of parts with close tolerances, precision welding, etc. The changeable sensor can be also either a vision camera, or a displacement sensor, or a proximity sensor. This invention is also applicable to precision teaching and positioning of robot's tools for precision assembling, welding, material removal, etc.