KR100473093B1 - Hoist system, method of improving responsiveness of hoist and system therefor, method of controlling hoist, control system for hoist and method of preventing slack in hoist line - Google Patents

Abstract

A human power amplifier assist device (10) includes an end-effector (14) tha t is grasped by a human operator and applied to a load (25). The end-effector (14) is suspended, via a line (13), from a take-up pulley (11), winch or dru m that is driven by an actuator (12) to lift or lower the load (25). The end- effector (14) includes a force sensor (31) that measures the vertical force imposed on the end-effector (14) by the operator and delivers a signal to a controller (20). The controller (20) and actuator (12) are structured in suc h a way that a predetermined percentage of the force necessary to lift or lowe r the load (25) is applied by the actuator (12), with the remaining force bein g supplied by the operator. The load (25) thus feels lighter to the operator, but the operator does not lose the sense of lifting against both the gravitation and inertial forces originating in the load (25).

Description

Hoist system, method and system for improving hoist responsiveness, hoist control method, hoist control system and slack prevention method of hoist cord {HOIST SYSTEM, METHOD OF IMPROVING RESPONSIVENESS OF HOIST AND SYSTEM THEREFOR, METHOD OF CONTROLLING HOIST, CONTROL SYSTEM FOR HOIST AND METHOD OF PREVENTING SLACK IN HOIST LINE}

This application is directed to U.S. Provisional Application No. 60 / 134,002, filed May 13, 1999, U.S. Provisional Application No. 60 / 146,538, filed July 30, 1999 and U.S. Patent Application, July 30, 1999 Priority is claimed on provisional application 60 / 146,541, which is incorporated herein by reference.

The present invention relates to a material handling apparatus for lifting and lowering loads by a worker's manpower.

The apparatus described herein is different from the conventional material handling apparatus used by automated assembly and warehouse workers. Conventional devices include three types of material handling devices currently available on the market.

A material handling device of the kind called balancer consists of an electric winding pulley, a string wound around the pulley as the pulley rotates, and an end-effector attached to the end of the string. The end effector has a part that is connected to the article to be pulled up. As the pulley rotates, the rope is wound or unrolled to raise or lower the load connected to the end effector. In this kind of material handling system, the actuator generates a pulling force on the string equal to the gravity of the object being pulled so that the tension is in line with the weight of the object. Therefore, the only force that the operator must apply for the movement of the object is the acceleration force of the object. This force can be quite large if the weight of the object is large. Therefore, acceleration and deceleration of heavy objects are limited by the force of the operator.

There are two ways to tension a string so that it is exactly balanced with the weight of the object. First, when the system is operated at pneumatic pressure, the air pressure is adjusted so that the pulling force is in balance with the weight of the load. Secondly, when the system is electrically operated, an appropriate amount of voltage is provided to the amplifier so that the pulling force is balanced with the weight of the load. The statically set tension of the balancer is not easy to change in real time, so this kind of system is not suitable for moving objects of varying weight. This is because each object needs a different deflection force to offset its gravity. This is cumbersome because the operator must manually adjust the weight of the object or weigh it. For example, pneumatic balancers made by Zimmerman International Corporation or Knight Industries are based on this principle. The air pressure is set and controlled by the valve to maintain a constant load balance. The operator manually accesses the actuator and sets the system to the desired pressure so that a constant tension is produced in the cord. Scaflia's LIFTRONIC system machine is also a balancer, but it works electrically. When the system picks up the load, the LIFTRONIC machine gives the rope the same pulling force in the opposite direction as the weight of the object caught. This machine is superior to Zimmerman pneumatic balancers because it has an electronic circuit that initially balances the load when it is picked up by the system. Thus, the operator does not have to approach the actuator and adjust the initial tension of the string. In such a system, the weight of the load is first measured by the force sensor of the system. During these measurements, the operator should not touch the load and instead allow the system to determine the weight of the object. If the operator touches an object, measuring the tension becomes inaccurate. This causes the LIFTRONIC machine to give the rope an unequal pulling force in the opposite direction to the weight of the object caught. Unlike the auxiliary device of the present application, the balancer does not allow the worker to physically feel the force required to lift the load. Also, unlike the device of the present application, the balancer only offsets the gravity of the object at the tension of the string and is not versatile enough to be used when the load is heavy.

The second type of material handling device is similar to the balancer described above, but the operator adjusts the rate of rise and fall of the object being moved using an intermediary device such as a valve, push-button, keyboard, switch or teach pendant. do. For example, if the operator opens the valve more, the lifting speed of the object becomes higher. By using the intermediary device, the operator does not have physical contact with the load to be raised, but is cumbersome to operate the valve or the switch. The worker does not know how much he raises because his hand is not in contact with the object. Although suitable for lifting objects of varying weights, this type of system is not convenient for the operator since the operator must concentrate on the intermediate device (ie valve, push-button, keyboard or switch). The operator therefore has to concentrate more on the operation of the intermediary device than on the speed of the object, which makes the impression manipulation unnatural.

The third kind of material handling apparatus uses an end effector with a force sensor or a motion sensor. Such a device measures the force or motion of an operator and changes the speed of the actuator based on it. An example of such a device is US Pat. No. 4,917,360 to Yasuhiro Kojima. By using such and similar devices, the pulley rotates to raise the load when the operator pushes the end actuator upwards, and the pulley rotates to lower the load when the operator pushes the end actuator downward. Problems arise when the operator pushes the end effector downwards to engage the suction cup, which the controller and actuator understand this to lower the load. This causes the actuator to rotate to loosen the line more than necessary, causing "slack" in the line. Hereinafter, the term "slack" is understood to be the excess length of the string, but not to be construed to include the case where the string is not only fully taut. Slack cords can be wound around the operator's neck or hands. After slack has occurred in the rope due to this or other circumstances, the worker may press the handle upwards and the strap may be wound close to the operator's neck or hand to cause fatal injury. Slack may occur even when a suction cup is not used as a means of holding a load, and it is important to prevent slack at all times for safety. During fast movement, the operator may accidentally lift the load or touch the bottom of the end effector to place it on a peripheral device (eg conveyor belt). In pallet work, the operator uses the bottom of the end effector very often to finely adjust the position of the box that is difficult to place in place. This phenomenon causes a slack in the string because the operator pushes the end actuator downward to adjust the position of the box, where the end actuator is restricted from moving downward. In general, slack in strings can be dangerous for workers and other identical working environments. In the manual material handling apparatus of the present invention, no slack occurs in the string.

This type of force sensor device also does not allow the operator to actually feel the weight of the load being lifted. This can make the movement of goods unnatural and dangerous.

1 shows an embodiment of an attraction amplifier comprising an end effector according to the invention.

FIG. 2 is a cross-sectional view of one embodiment of an end effector usable in the present invention, showing a particular structure of a force sensor measuring force of an operator. FIG.

3 is a cross sectional view of one embodiment of another end actuator including a displacement detector for measuring a force exerted on the end actuator by an operator;

4 is a perspective view when the end effector of FIG. 3 is used to lift a box by an operator;

5 is a block diagram showing the force and load of a worker interacting with a component of the attraction amplifier to move the load.

FIG. 6 illustrates the problem of slack in strings that may occur in the prior art using suction cups to grasp the box.

FIG. 7A is a partial cross-sectional view of one embodiment of an end actuator including a displacement detector for measuring the force exerted by the operator on the end actuator and a force sensor for measuring the tension of the string; FIG.

FIG. 7B is a partial cross-sectional view of one embodiment of an end actuator including a displacement detector for measuring the force exerted on the end actuator by a worker and a force sensor for measuring the force associated with the weight and acceleration of the load only; .

8 schematically illustrates the use of a force sensor that can be used to measure the total force exerted on a ceiling or overhead crane by the attraction amplifier.

9 shows schematically an embodiment of an actuator comprising a mechanism for measuring the tension of a string and a motion sensor.

10A and 10B are partial cross-sectional views illustrating one embodiment of an end actuator including a displacement detector for measuring the force exerted by the operator on the end actuator and a mechanism for detecting the tension of the string.

11A and 11B illustrate an embodiment of an actuator including a switch and a mechanism for detecting the tension of a string.

12A and 12B illustrate an embodiment of an end actuator comprising a displacement detector for measuring a force exerted on the end actuator by an operator and a switch that transmits a signal when the end actuator is constrained to move downward; Drawing showing.

FIG. 13 illustrates the use of a clamp-on current sensor that can be used to detect the current consumed by an actuator. FIG.

FIG. 14 is a block diagram showing the force and load of a worker interacting with a component of the attraction amplifier to move the load while the slack of the string is prevented.

15A, 15B, 15C, and 15D are graphs showing control variable K _M values as a function of tension in hoist strings.

FIG. 16 illustrates one embodiment of an attraction amplifier that prevents slack in the string even when the end actuator is pressed down by an operator while the end actuator is restricted from moving downward;

FIG. 17 is a block diagram showing the force and load of an operator used as a feedback signal for moving the load while the slack of the string is prevented.

18A and 18B are flowcharts of software that can be used to drive a controller in practicing the present invention.

The auxiliary device of the present invention solves the problems described above with respect to three kinds of material handling devices. The hoist of the present invention is designed to provide an end tension between the operator, an actuator such as an electric motor, a computer or other type of controller for controlling the actuator, and a tension between the cord, cable, chain, rope, wire or actuator and the end actuator. Contains a different kind of line for passing. Hereinafter, the term "impression" should be understood to include the movement of both the up and down of the article. The end effector provides an interface between the operator and the object to be raised. Force transmission mechanisms, such as pulleys, drums or winches, are used to apply the force generated by the actuator to the string that transmits the pulling force to the end effector.

A signal indicative of the vertical force exerted on the end effector by the operator and measured by the sensor is transmitted to a controller associated with the actuator. In operation, the controller causes the actuator to properly rotate the pulley and move the end effector so that the operator only lifts a small, pre-programmed portion of the load while the remaining force is provided by the actuator. The actuator therefore assists the operator during the hike movement in response to the force of the operator's hand. The tension of the cord is also detected or measured, for example by the energy or current consumed by the actuator. Also, because load loading is the main factor in determining the magnitude of tension, load loading can be used to approximate the tension and vice versa. Hereinafter, it should be understood that the tension can be calculated using the load and the load can be calculated using the tension. A signal indicative of the load or string tension is transmitted to the controller, which uses the load load signal or tension signal to effectively drive the actuator according to the operator's input. This can, for example, prevent the actuator from releasing the string in the event of a zero load or tension, so that no slack (as defined above) occurs on the string even if the string is loose (i.e. not taut). do.

By this bag sharing concept, the worker has a feeling of lifting the load but is much smaller than the normally required force. The force exerted by the actuator takes into account both the gravity and the inertia forces required to move the load. Since the force exerted by the actuator is automatically determined by the force of the joule and the force exerted on the end effector by the operator, there is no need to set or adjust the attraction amplifier for loads of different weight. The attraction amplifiers do not have switches, valves, keyboards, teach pendents or push-buttons to control the lifting speed of the load. The contact force between the end effector handle coupled with the operator's hand and string force is used to determine the speed of pulling up the load. The force of the operator's hand is measured and used by the controller in cooperation with the force of the rope, giving the required angular velocity of the pulley to raise and lower the rope, thereby providing sufficient mechanical strength for the worker working on the lifting. In this way, the device follows the operator's arm movement in a "natural" manner. When an operator uses these devices to handle a load, an appropriately defined portion of the total load (sum of gravity and acceleration) is lifted by the operator. This force allows the operator to feel how much weight is lifting. Conversely, if the operator does not apply any vertical force (up or down) to the end effector handle, the actuator does not rotate the pulley at all and the load hangs without moving on the pulley.

Although the conventional apparatus described above raises a load, these apparatuses

Do not let the worker feel the impression movement physically,

Does not compensate for inertia,

No compensation for various loads,

Does not address ergonomic concerns,

It does not prevent slack in the line.

The device of the present invention does not have the above disadvantages.

1 shows a first embodiment of the present invention, which shows a human power amplifier 10. A winding pulley 11 driven by an actuator 12 is mounted directly on the ceiling, wall or overhead crane at the top of the device. A string 13 is wound around the winding pulley 11. When the pulley 11 is rotated, the rope 13 can raise or lower the article 25. The string 13 can be wound around the pulley to provide the force to raise the load, which can be any type of wire, wire, cable, belt, rope, wire string, cord, twist, string or other member. have. The row 13 has an end-effector 14 comprising a human interface subsystem 15 comprising a handle 16 and a load interface subsystem 17. Is attached, and in this embodiment this end effector 14 comprises a pair of suction cups 18. Moreover, the suction cup 18 is provided with the air hose 19 which supplies low pressure air.

In a preferred embodiment, the actuator 12 may be replaced by an electric motor with a transmission or an electric motor without a transmission. In addition, actuator 12 may be powered using other types of energy, including pneumatic, hydraulic, and other alternative forms of energy. As used in this embodiment, the transmission is a mechanical device such as a pulley and a string that increases or decreases the tensile force of the string. The pulley 11 may be replaced by a drum, winch or other mechanism capable of converting the movement provided by the actuator 12 into a vertical movement that raises or lowers the rope 13. In this embodiment, the actuator 12 powers the winding pulley 11 directly, but the winding pulley 11 is mounted at another position of the actuator 12 and powered through other transmission systems such as chain and sprocket assemblies. It may be delivered to the winding pulley (11). The actuator 12 is driven by the electronic controller 20 which receives a signal from the end effector 14 via the signal cable 21. Since there are several ways of transmitting electrical signals, the signal cable 21 can be replaced with other alternative signal transmission means (e.g. RF, light, etc.). In a preferred embodiment, the controller 20 essentially includes the following three main components.

1. Computers, analog or digital circuits with input / output capability and standard peripherals. The controller's function in this area is to process the information received from the various sensors and switches to generate a command signal to the actuator.

2. A power amplifier that transmits power to the actuator based on the computer's instructions. Typically, this power amplifier receives power from a power supply and delivers the right amount of power to the actuator. The amount of power supplied by the power amplifier to the actuator 12 is determined by the command signal calculated by the computer.

3. Logic circuit consisting of an electromechanical relay or a solid state relay that starts and stops the system according to a series of possible events. For example, a relay uses two push buttons installed on a controller or end actuator to start and end the operation of the entire system. This relay engages a friction brake in the event of a power failure or operator left the system.

The operator interface subsystem 15 is designed to measure the force of the operator, that is, the force exerted on the operator interface subsystem 15, by the operator. The load interface subsystem 17 is designed to be in contact with the load and to accommodate various holding devices. The design of this load interface subsystem depends on the load geometry and other factors related to the pulling behavior. In addition to the suction cup 18 shown in FIG. 1, hooks and grippers connected to the load interface subsystem are examples of other means. When lifting a heavy object, this article interface subsystem may include two or more suction cups.

The operator interface subsystem 15 of the end actuator 14 includes a sensor (described below) that measures the amount of vertical force exerted by the operator. If the operator's hand pushes the handle 16 upwards, the winding pulley 11 moves the end actuator 14 upwards. If the operator's hand pushes the handle 16 downward, the take-up pulley moves the end effector 14 downward. The measured hand force of the operator is transmitted to the controller 20 via the signal cable 21 (or other replaceable signal transmission means). The system of the preferred embodiment of the present invention also includes a sensor disposed proximate to the end actuator 14 while an operator-applied force that is not proximate to the end actuator 14 for estimating operator attraction. Estimation components may be used.

Using these measuring devices, the controller 20 assigns a pulley speed necessary to raise or lower the string 13 which generates enough mechanical force to assist as necessary for the operator's pulling operation. The controller 20 then supplies power to the actuator 12 via the power cable 23 to cause the pulley 11 to rotate. All of these actions take place so quickly that the operator's attempts to raise and the device's attempts to happen at the same time occur completely simultaneously. Thus, the physical movement of the operator is changed to the physical assistance of the mechanism, and the force of the mechanism is controlled directly and simultaneously by the operator. In summary, the load moves vertically because both the worker and the pulley move vertically. One of the most important characteristics of the device according to the invention is that the end effector rotates the actuator and pulley according to the hand movement of the operator up and down, and the movement of the end effector downward is restricted and the end effector is moved downward by the operator. The line does not loosen even if it is pushed.

A dead-man switch with a lever 26 on the handle 16 (described below) sends a signal to the controller 20 via a signal cable 22 (or other alternative signal transmission means). do. When the operator is holding the handle 16, the deadman switch transmits a logic signal to the controller 20 that causes the end effector to move along the operator's hand. When the operator releases the handle 16, the deadman switch transmits another logic signal to the controller 20 that causes the end effector to remain stationary. In a preferred embodiment of the invention, a friction brake 24 is installed on the actuator 12. This friction brake always engages when the operator releases the deadman switch or a power failure occurs. It is possible to use an end effector with two handles, but only one of them needs to be installed as a sensor for measuring the operator's action. When lifting a heavy object, two attraction amplifiers similar to the attraction amplifier 10 shown in FIG. 1 can be used, one for the left hand and the other for the right hand.

First, the architecture of two types of end effectors will be described in detail in consideration of the force measurement of the operator. The control algorithm then considers the operation of the system and the prevention of line slack. A flowchart is also given to illustrate the implementation of this control algorithm.

Two types of end effectors are described here. 2 is an example of an end effector 31 which measures the force of a person via a force sensor. The force sensor 31 is installed between the handle 32 and the bracket 33 and is connected to the controller 20 via the signal cable 21. The force sensor 31 has a threaded portion 34 that is screwed into the bore in the handle 32 that the operator grabs. The opposite side of the force sensor 31 is connected to the bracket 33 via the cylinder 35. The outer diameter of the cylinder 35 is slightly smaller than the inner diameter of the handle 32. This clearance enables the sliding action between the handle 32 and the cylinder 35 and is transmitted to the force of the operator in the vertical direction and to the bracket 33 and transmitted to the force sensor 31 without any resistance. The sensor 31 ensures the operator's force in a non-vertical direction that is not transmitted. If a force other than the vertical direction is transmitted to the force sensor 31, it may cause a false reading of the sensor or damage of the force sensor assembly.

Force sensors used in embodiments of the present invention include piezoelectric based force sensors, metal strain gauge force sensors, semiconductor strain gauge force sensors, Wheatstone bridge built strain gauge force sensors, and force sensing resistors. You can choose from a variety of commercial force sensors, including. Regardless of the particular type of force sensor selected and its installation process, the force sensor 31 should be designed to measure only the force exerted by the operator on the end actuator 30. The bracket 33 is rigidly connected to the cylinder 35 and includes an eyelet 37 which is connected to the hook 13 and the string 13 in contact with the load.

In the second group of embodiments, the force exerted by the operator on the end effector is measured by the displacement of the handle rather than the force sensor of the kind described above. The low cost and ease of use of the displacement measuring system make this type of end effector in some cases even more attractive. A partial cross sectional view of one embodiment of this second group of end effectors is shown in FIG. 3. FIG. 4 shows a perspective view of the end effector of FIG. 3 when the operator uses it to lift the box 25. Similar to the end effector described above, the end effector 40 includes an operator interface subsystem 41 and a load interface subsystem 42. The operator interface subsystem includes a handle 16 that the operator grabs and thus measures the force of the operator rather than the load.

The load interface subsystem 42 includes a bracket 44 that is bolted to a hook 45, a suction cup or other type of device capable of holding an object. Eyelets 46 are mounted to brackets 47 to connect brackets 47 to strings 13.

The handle 16 is held by the operator and firmly connected to the ball screw portion 49 of the ball spline shaft mechanism. The ball 50 located in the groove of the spline shaft 51 allows the ball screw 49 and the handle 16 to freely move along the spline shaft 51 without rotation associated with the spline shaft 51. The spline shaft 51 is secured to a bracket connected to the string 13 via eyelets 46.

In this embodiment, the spline shaft 51 is press-fitted to the bracket 47. The bracket 44 supporting the hook 45 is connected to the bracket 47 with bolts 52. The bracket 44 has a hole pattern to which a suction cup mechanism, a hook or other device for holding an object can be connected. The coil spring 53 is located around the spline shaft 51 between the ball screw portion 49 and the stopper 54 of the ball spline shaft mechanism and directs the handle 16 upwards. It should be noted that the stopper 54 may be a clamp ring firmly fixed to the spline shaft 51.

In this embodiment, the linear encoder measures the movement of the handle 16 relative to the bracket 47. This encoder system comprises a sensor 48 for generating an electrical signal on the signal cable 21. This encoder has a reflective strip 55 attached with an adhesive on the handle 16. This reflective strip has dark horizontal lines. As the handle makes a linear movement relative to the bracket 47, the sensor 48 detects bright and dark areas of the strip 55, and detects the bright (or dark) areas of the strip 55 as appropriate pulses. Is transmitted through the signal cable 21. The rising and falling edges of the pulse signal are counted at the controller 20. FIG. 3 shows the end effector when the handle 16 is pushed upwards to its upper limit (the ball screw portion 49 presses the bracket 47). It is better to laser mark the handle 16 itself than to glue the reflective strip with dark stripes onto the handle 16. The controller assumes the zero position of the handle 16 at this position and counts the pulses transmitted through the signal cable 21 to calculate the handle displacement. Both the handle displacement and the spring stiffness are considered to calculate the value for the operator's force. The linear motion detector used in this embodiment is a magnetic linear encoder, a linear potentiometer, a linear variable differential transformer (LVDT), a capacitive displacement sensor, an eddy current proximity sensor, or a variable inductance. (variable-inductance) proximity sensor.

Alternatively, the ball spline shaft mechanism shown in FIG. 3 can be replaced by a linear bushing mechanism, with the bushing (slider) and the shaft sliding against another without the ball. There is some friction between this bushing and the shaft.

Deadman switch 56 installed on handle 16 transmits a signal to controller 20 via signal cable 22 (or other alternative signal transmission means). A lever 26 pivoting around the hinge 58 is installed on the handle 16 and presses the switch 56 when the operator grasps the handle 16. In a preferred embodiment of the present invention, the friction brake 24 is mounted on the actuator 12. This friction brake engages whenever the operator releases the deadman switch and in the event of a power failure. In addition, as an optional feature, an auxiliary device controller can be designed that causes the controller to put the actuator into position control mode when the operator releases the handle 16. In the position control mode, the controller keeps the actuator (and consequently the end actuator) in its position where the operator has left it. As soon as the operator returns and grabs the handle 16, the actuator leaves the position control mode. In a preferred embodiment, the position control mode includes a standard feedback system that maintains the position of the actuator in a position the operator has left using an encoder on the actuator in accordance with the feedback signal. Although this optional feature keeps the actuator and end actuator stationary when the operator releases the handle, it is not recommended to those skilled in the art to replace the previously described friction brake with this feature. The position control feature will not work if there is a sensor, computer or power failure.

The sole purpose of the spring mounted on the end effector is to return the handle position to the equilibrium position when the operator has not applied the handle. 3 shows an end effector using a compression spring. Other types of springs can be used for the end effector, such as cantilever beam springs, tension springs or disc springs. Basically any elastic element that can return the handle to the equilibrium position is enough. For example, a bellow can be used that not only protects the end actuator from dust and moisture, but also returns the handle to the equilibrium position. Structural damping of the elastic element (eg spring) or friction of the moving element (eg bearing) of the end effector provides sufficient damping to the system to impart stability. As shown in Figure 3, only one spring is used to push the handle upwards. However, two springs can be used to keep the handle in the center position. This second spring can be located around the spline shaft 51 between the ball screw portion 49 and the bracket 47 of the ball spline shaft mechanism, with the handle 16 facing upwards. As shown in FIG. 4, an optional brace 59 may be connected to the handle 16 to give the operator stability and convenience. This brace 59 has a hinge 57 and enables rotational movement along the arrow 43. Since the brace 59 transmits all the force applied to the hand of the worker through the wrist of the worker and below the arm of the worker, the worker can see that the brace 59 makes the work more comfortable.

As described above, other types of operator input estimation elements can be used in place of the above embodiments. Alternative examples of operator input estimation elements may include sensors that obtain the energy consumed by the actuator during pulling, or sensors that can estimate the load to load or the tension to estimate the operator's action without being close to the end actuator.

The block diagram of FIG. 5 shows the basic control technique of the device. As described above, in a preferred embodiment the force sensor or displacement sensor of the end effector transmits a signal to the controller 20 which is used to control the actuator 12 and apply the appropriate torque to the pulley 11. If (e) is an input command to the actuator 12, since there are no other external torques in the actuator, the linear velocity at the outermost point of the pulley or the velocity v of the end actuator can be expressed as:

Where (G) is the actuator transfer function. A positive (v) value means the velocity below the end effector. And the input command (e), joule tension (f _R ) from the controller contributes to the end effector speed as follows.

Where (S) is the actuator sensitivity transfer function associated with the joule tension (f _R ) for the end effector speed (v). If a closed loop speed controller is designed for the actuator such that (S) is small, the actuator has only a small response to joule tension. The high gain controller of the closed loop speed system results in a small (S), resulting in a small change in the speed (v) in response to the wye tension. In addition, non-back-driveable speed reducers (typically high transmission ratios) generate a small (S) for the system.

Joule tension (f _R ) can be represented by Eq.

Where (f) is the operator action on the end effector, and force (p) is the force applied by the load and the end effector here referred to as the "load load" on the string. Positive values (f) and (p) represent the downward force. Note that (p) is the force applied to the string, taking into account both the weight and the inertia forces of the load and end actuator.

Where W is the total weight of both the end effector and the load, Is the acceleration of the end effector. (P) is equal to the weight of the end effector and the load if the end effector and the load do not have the desired acceleration or deceleration. 5 and 4, the variable (E) in the block diagram of FIG. Care should be taken to show that the Therefore p = W-Ev.

The force f of the operator is measured and transmitted to the controller 20 which generates the output signal e. On the computer, subtract a positive number f _up from the measurement of the operator's force f. The role of f _up is described below. If the transfer function of the controller is represented by (K), the output (e) of the controller is as follows.

Substituting (f _R ) and (e) of equations (3) and (5) into equation (2) yields the following equation for the end effector speed (v).

The operator's upward force on the end effector can only be measured when the string is tensioned by the weight of the end effector. If the end actuator is light, the upward force of the full range of operators may not be measurable by the sensors of the end actuator. To solve this problem, a positive number f _up is introduced in equation (5). As shown in equation (6), the absence of (f), (p) and (f _up ) moves the end effector upwards. The maximum downward force applied by the operator is assumed to be f _max . Then it is preferable to set (f _up ) to about half of f _max . Equation (7) substituted with (f _up ) in equation (6) represents the article speed.

When the operator pushes downward so that f = f _max , the maximum descending speed of the end effector is

If the operator does not press at all, the maximum ascent rate of the end effector is:

Therefore, it is not necessary to worry about the measurement of the upward force of the operator by introducing (f _up ) into equation (5). If S = 0, the upward and downward maximum speeds are the ideal magnitude. However, if S, other than zero, appears under the same conditions as a given load, the magnitude of the maximum upward velocity is less than the magnitude of the maximum downward velocity. This is very natural and intuitive for the worker.

Returning to equation (6) again, it can be seen from equation (6) that the greater the force the operator exerts on the end effector, the higher the cargo speed will be. Using the measured operator force, the controller properly sets the pulley speed to generate enough mechanical force to assist the operator in lifting. In this way, the end effector "naturally" follows the arm movements of the operator. In other words, the pulleys, strings and end actuators mimic the operator's lifting / lowering movements, and the operator can more easily manipulate heavy objects without using other intermediary devices.

Now, some important characteristics of the device will be explained through three experiments. Substituting p into equation (6) and rearranging terms yields equation (10).

Equation (11) shows that the change in load weight (ΔW) and the change in force (Δf) that the operator exerts on the end actuator result in a speed change (Δv) of the end actuator.

Experiment 1

If Δv = 0 of two different objects in motion, that is, the operator maintains a similar operating speed,

Arrange the terms in equation (12) again to get equation (13).

Equation (13) indicates that if the work speed is not expected to change, an increase or decrease in weight will result in an increase or decrease in the upward force of the operator. In other words, as the load weight increases, the operator must increase his upward hand force or reduce the downward force to maintain the same operating speed. In equation (13), the term (GK / S + 1) is the force amplification factor. Choosing a larger (K) will amplify the greater force in the system. As a result, when the force amplification is large, the operator "feels" only a small percentage of the load weight change. Inherently, the load is still "lightly felt" inherently lighter if the operator continues to maintain a sense of the dynamic characteristics of the free mass. This baggage distribution method provides a sense of how much weight the operator is lifting. Looking at Eq. (13), it can be seen that if (S) is a small amount, the change in load weight (ΔW) produces a small change in operator force (ΔW). In other words, if (S) is a small amount, the operator will feel less about load weight changes. How to solve this problem will be described later, and when (S) is a small amount, the operator will feel the change of load more clearly. It should also be noted that at very low frequencies (slightly slow and gentle movement) the left side of equation (13) approaches a large number. This indicates that if the operating speed is not expected to change, the increase or decrease of the load weight (ΔW) increases or decreases the operator's upward force very little (not noticeably). However, at high frequencies (slightly fast and rough movement), the operator will feel the weight change more noticeably. In other words, if the worker performs a relatively slow pulling exercise, the additional force required to maintain the heavier load speed for lighter loads will not be noticeable. However, if the worker performs a quick pull exercise, the additional force needed to maintain the heavier load speed for lighter loads will be more noticeable.

Experiment 2

If Δf = 0 (that is, the operator decides to maintain a similar force on the end effector for two different weights), Eq. (11) can be reduced to

This means that if the operator maintains a constant hand force, the increase in load weight (ΔW) results in an increase in the descending speed. Furthermore, an increase or decrease in load weight (ΔW) will result in a decrease or increase in the ascending speed of the end actuator relative to the operator's force given to the end actuator, respectively. In essence, if there is an increase in the weight of a load for a given operator's force, the load will descend faster and lift more slowly. If the load weight increases from equations (13) and (14), the operator needs to either increase the upward force to maintain the same operating speed or decrease the upward operating speed to maintain the same operator's own hand force. Can deduce. This dynamic action is very comfortable and natural for the operator.

Experiment 3

Finally, ΔW = 0, i.e. the load weight is constant:

This means that if the weight of the load does not change, an increase or decrease in the downward force of the operator (Δf) results in an increase or decrease in the lowering operating speed, respectively. Equation (15) can also be interpreted differently, where an increase in operating speed for a given load weight requires a larger operating force. In general, the larger (K) the smaller the operating force will be.

As shown in Fig. 5, (K) cannot be arbitrarily large. Rather, the choice of (K) should ensure the closed-loop stability of the system shown in FIG. Since the operator's force (f) is the impedance function (H) of the operator's arm, the load (p) is a function of the load dynamics (E), i.e. the weight and inertial forces produced by the load. There are many ways to design a controller transfer function (K). A paper by Kazerooni and Snyder entitled "A Case Study on Dynamics of Haptic Devices: Human Induced Instability in Powered Hand Controllers" describing the conditions of closed loop stability of the system [AIAA Journal of Guidance, Control, and Dynamics, Vol. . 18, No. 1, 1995, pp. 108-113, which is incorporated herein by reference and forms a part of this specification. Those skilled in the art are not limited to the choice of controller, and in many cases a simple low pass filter with the controller is suitable for system stabilization of FIG. Several choices for linear or nonlinear controllers can lead to overall performance improvements (large force amplification and high speed operation) before changes in operator arm impedance (H) and load dynamics (E).

The choice of (K) also depends on the effective computational power: high speed computer and large memory resulting in sophisticated control algorithms that stabilize the closed system of FIG. 5 and large power amplification with high speed movement. Need. A paper by H. Kazerooni and J. Guo entitled "Human Extenders" describing the stability of closed-loop systems and (K) design method [ASME Journal of Dynamic Systems, Measurements, and Control, Vol. 115, No. 2 (B), June 1993, pp. 281-289, which is incorporated herein by reference, forms a part of this specification.

Theoretical (G) and (S) values can be obtained using standard modeling techniques. There are many experimental frequency- and time-domain methods for measuring (S) and (G) that produce better results. It is recommended to use the frequency domain technique to identify (G) and (S). For example, a book titled "Feedback Control of Dynamic Systems" by G. Franklin, D. Powell and A. Emami-Naeini, Addison Wesley describes in detail the frequency and time domain methods of demonstrating various transfer functions. do.

Note that the linear system theory is used in this embodiment to model the dynamic behavior of system components. This allows for the disclosure of system features in the simplest and most commonly used form. Most skilled artisans will be familiar with linear system theory and will be able to understand the basic theory of the present invention using the mathematical tools of linear system theory (ie transfer function). However, nonlinear models can also be used and the mathematical procedures described above can be followed to account for the dynamic behavior of the system.

If the operator pushes the end actuator downward, special problems may arise in the apparatus but the end actuator is prevented from moving downward. This condition can be illustrated by the following example using a suction cup to grab the article. The end actuator 14 as shown in FIG. 6 has an actuator (not shown in FIG. 6) to secure the engagement of the box 25 and the suction cup 18 when the operator pushes the handle 16 downward. Will loosen the line (13). This is in response to the downward force of the operator acting on the end actuator 14, which transmits a command signal to the actuator which causes the operator to lower the end actuator so that the actuator releases the string 13 Is caused by misjudgement. In some instances, the loose "slack" portion of the string is several feet. After engaging the suction cup 18 with the box 25 attached, the operator pushes the handle 16 upward to raise the box, the slack of the string 13 must be treated before the actuator and the unwinding box are raised. . This hinders the operator because the operator has to wait while the actuator winds up the slack of the rope 13. In addition, the end actuator 14 will move suddenly if the tension of the string changes from zero (i.e. the string is loose) to a sudden non-zero value (i.e. the string is not loose). This sudden movement can drop the box. In short, the worker's movement during the pulling operation may be disturbed due to unnecessary slack of the rope 13 and the box may fall as the rope suddenly changes from a loose state to a tight state.

 String slack can have far more serious consequences than slowing a worker's work, such as a loose string wrapping the worker's neck or hands. As described above, when the worker pushes the handle upward after the slack is generated in the string, the string in the slack state may cause severe injury or fatal injury by tightening the operator's neck or hand. Therefore, it is important not to cause slack in the string 13.

According to another feature of the invention, the actuator does not loosen the string 13 when the operator pushes down the handle 16 of the end effector to strongly engage the suction cup 18 and the box 25. In other words, the device of the present invention has "intelligence" to recognize that the operator simply pushes down to engage the box to the suction cup 18 and that the operator does not intend to move his hand further down. On the other hand, when the operator pushes the end actuator handle 16 downward, the actuator of the present invention unscrews the string 13 so that the movement of the end actuator is not affected when there is no box, thereby preventing the worker movement downward. I will not accept. The above-described auxiliary device can be distinguished into two cases, namely, the case in which the actuator does not loosen the string 13 and the case in which the actuator loosens the string 13.

In order to prevent slack of the string 13, the tension f _{r of the} string must be detected. Then, the pulley speed should be adjusted with the tension of the known rope so that the rope does not loosen unnecessarily, thereby preventing slack in the rope. The simplest form to prevent line slack is to stop the actuator and pulley when (f _R ) reaches zero. A more complex form to prevent string slack smoothly requires the pulley's rotational speed to approach zero when the tension (f _R ) of the rope approaches zero, and a zero-tension limit is registered in the controller of the rope. The pulley rotation speed to zero. In other words, if the tension is zero, the pulley speed is properly reduced to zero to prevent string slack.

Earlier, it was described that the pulley speed depends on a signal indicating only the force of the operator. However, in the case of a device in which no string slack will occur, it relies on a signal indicating the tension of the string in addition to the signal indicating the operator's force on the end effector handle. The following two methods are preferable for detecting the tension of the string. The first method is to directly detect the tension, and the second method is to measure the tension of the string by measuring the power or current consumed by the actuator and estimating the tension of the string based on it. Known string tension can then be used to bring the actuator and pulley to zero speed to prevent string slack.

If the tension of the string is detected directly, the force sensor can be used to directly measure the tension of the string. FIG. 7A shows the end actuator 60 with a force sensor 61 installed on the end actuator between the end actuator 60 and the string 13. The force sensor 61 is attached to the bracket 47 of the end effector using the screw 62. Connect the bracket to the force sensor 61 using a set of screws 63. Eyelet 16 is screwed to bracket 64 and provides an interface with string 13. Since the force between the string 13 and the end effector 6 is transmitted to the force sensor 61, the force sensor 61 always measures the tension of the string. The signal cable 65 transmits a signal indicating the tension of the string to the controller 20.

Alternatively, a force sensor can be installed on the end effector to measure only the force associated with the load as shown in FIG. 7B. The force sensor 61 is connected to the bracket 44 with screws 62. Connect the bracket 64 to the force sensor 61 using a set of screws 63. The suction cup 18 is connected to the bracket 64 and provides an interface with the box 25. In this case, the force sensor 61 always measures the same force as the inertial force due to the weight and acceleration of the article only. Since the signal cable 65 transmits a signal representing this force to the controller 20, the controller can know the force (indicated by p _L ) represented by the weight of the load and the inertia force. The measurement of p _L and f together with the calculation of the end effector acceleration (or direct measurement) leads to the calculation of the tension of the string according to equation (16) below.

Where W _E is the weight of the end effector itself and is known in advance. When traveling at low accelerations, the force measured by the sensor is always tension (e.g. positive value), unless the string is slack. The moment the load meets the obstruction blocking the downward movement, the sensor shows a compressive force (eg a negative value). This change in sign while measuring p _L indicates that there is a state of zero tension in the string. In addition, since the load load p _L is usually larger than the operator's operating force f, the tension f _R can be roughly estimated by ignoring f in equation (16). After all, when moving at low acceleration, the tension of the string is approximately equal to the sum of the weight of the end actuator and the weight of the load. Here, those skilled in the art should assure that the tension of the string is satisfied of equation (16) using any signal representing zero.

Force sensors suitable for use in the present invention can be selected from a variety of commercial force sensors including piezoelectric based force sensors, metal strain gauge force sensors, semiconductor strain gauge force sensors, strain gauge built-in Wheatstone bridge strain gauge force sensors, and force sensing resistors. Regardless of the particular type of force sensor selected and its installation process, the force sensor must be designed to estimate load or tension with reasonable accuracy.

Alternatively, as shown in the attraction amplifier 70 of FIG. 8, the force sensor can be installed directly between the actuator 12 and the rail or trolley. The force sensor 71 measures the total force exerted by the pulling device on the rails 72. The signal indicative of this measured force is transmitted to the controller 20 via the signal cable 73. If the tension is zero, the output signal of the force sensor indicates the weight of all components connected to the actuator, pulley, brake, and rail 72. This value can be measured in advance and stored in the controller memory. If the tension of the string is not zero, the output signal of the force sensor increases including the tension of the string. Therefore, the tension of the string can be detected by subtracting a constant value (stored in memory) from the output signal of the force sensor.

9 shows how a motion sensor or estimator is used to measure string tension. The string 13 is wound around the pulley 11 and the actuator 12 is connected to the trolley 81 by bolts 82. Bar 83 freely rotates around point 84 on the actuator body, holds idle pulley 85 on one end, and connects tension spring 86 on the other end. . Tension spring 86 is secured to the actuator body at point 87. The idle pulley 85 pushes the string 13 by the force of the spring 86. The rotation of the bar 83 is measured with an angular motion sensor 88. Various motion sensors can be used, such as optical encoders, resolvers, or potentiometers, to measure the rotation of the bar 83 relative to the actuator body. If the string tension is greater, the bar 83 rotates more counterclockwise. When the tension of the string is a small value, the bar 83 rotates clockwise by the force of the tension spring 89. The signal cable 89 delivers the output of the motion sensor 988 to the controller. The output signal of the motion sensor 88 can be adjusted to measure or estimate the tension value of the string. This is accomplished by installing an idle pulley that translates on the bar. At this time, a linear encoder or LVDT can be used to detect this linear motion.

Instead of generating a signal indicating the magnitude of the tension in the string, one of the signals is a signal indicating that the tension in the string is zero, and the other signal is of interest to a detection device that generates a binary signal indicating that the tension in the string is not zero. There may be. The device is cheaper because it provides limited information about the tension of the string. 10A and 10B show an end effector 90 having a tension detector that includes a momentary switch 91 for generating a binary signal installed on the bracket 47. The string 13 is firmly connected to the bracket 47, and the plate 92 can rotate along the hinge 93. The tension spring 94 is connected between the plate 92 and the bracket 47 which causes the plate 92 to rotate in the direction of the arrow 95. The plate 92 also has a hole through the string 13. Signal cable 96 delivers the output of the momentary switch to the controller. The stopper 97 is preferably a plastic sphere and is firmly connected to the string 13. The stopper 97 prevents the plate 92 from rotating along the direction of the arrow 95 when the string tension is not zero (FIG. 10A). In fact, if the tension of the string 13 is found to be non-zero, the stopper 97 causes the plate 92 to be in the position shown in FIG. 10A where the switch 91 does not push the switch 91. If the tension of the string is zero (FIG. 10B), plate 92 pushes switch 91 with the force of spring 94. Thus, this limited force detection device detects the tension present in the string but cannot measure the magnitude of the tension in the string. Basically, this method generates a string tension, a binary electrical signal indicating the presence or absence of tension on the string, one indicating that the tension of the string is not zero, and the other signal to the controller indicating that the tension of the string is zero. It is used to make.

Alternatively, one may be interested in using the tension of the string at other locations on the string to detect the presence of the tension of the string. This is illustrated in FIGS. 11A and 11B, where a binary signal is generated to the controller using the tension of the string at the top of the device near the actuator 12. The string 13 is wound around the pulley 11 and the actuator 12 is connected to the trolley 81 by bolts 82. The bar 83 rotates freely along the point 84 on the actuator body, holds the idle pulley 85 on one arm and connects the tension spring 86 on the other arm. Tension spring 86 is secured to the actuator body at point 87. The idle pulley 85 pushes the string 13 by the force of the spring 86. If the tension of the string is not zero as shown in FIG. 11A, the tension of the string exceeds the spring force to cause the bar 83 to fall from the switch 98. When the tension of the string is zero as shown in FIG. 11B, the idle pulley 85 is pushed to the left by the force of the tension spring 86. This causes switch 98 to be activated by bar 83. Thus, if the string tension is zero, the switch generates a signal. The signal cable 99 delivers the output of the momentary switch to the controller. Instead of converting the tension into the rotational motion as shown in FIGS. 11A and 11B, the tension of the string may be converted into the linear motion. This can be accomplished by doing translational movement and installing the idle pulley 85 on a bar supported on the linear bearing. The idle pulley is in contact with the string 13 and the tension of the string causes the bar to translate. The movement of the bar activates the switch 98 in response.

Another preferred method of detecting the tension state of the strings comprises end effector 79 means with a switch as shown in Figs. 12A and 12B. The switch 74 is preferably installed in the horizontal portion of the bracket 44. The bracket 75 holding the two suction cups 18 slides freely in the vertical direction with respect to the bracket 44. FIG. 12A shows the end actuator 79, where the end actuator is not restricted downward by any object and the switch 74 is not pressed. An optional compression spring 78 is installed between the bracket 75 and the bracket 44 to maintain the distance between the bracket 44 and the bracket 75. When the end effector is lowered (FIG. 12B), the bracket 75 is prevented from moving downward by the box 25, which causes the bracket 75 to press the switch 74 to generate an electrical signal. At this time, the total force associated with the weight and inertia of the end effector and the operator's force (right hand side of Eq. 16) are supported by the box 25 and not by the string 13. This means that the string tension (left side of equation 16) is zero. Thus, the signal generated by switch 74 determines not only the presence of obstructions, but also the presence of zero joule tension. Therefore, the sensor system of FIGS. 12A and 12B is not only an obstacle detector but also a tension detector. This signal can be transmitted by the signal cable 77 to the controller 20 to indicate that the tension of the cord is zero. If no object obstructs the downward movement of the end effector, the bracket 75 is lowered down by its own weight or by the force of the compression spring 780. Thus, this end effector can generate a binary signal, One occurs when the tension of the string is zero, and the other when the tension of the string is not zero.

The second preferred method estimates the tension of the string based on the current or energy consumed by the actuators supporting the end actuators and any articles connected on the string. The energy consumed by the system to support the end effector and the articles connected thereto may include several different forms of energy, including electrical energy, pneumatic energy, hydraulic energy and other alternative forms of energy. If pneumatic or hydraulic actuators are used in the system, the load pressure of the actuator can be used to estimate the tension of the cord. In a particular preferred embodiment, the tension of the string can be determined by measuring the current of the electric actuator since the current of the electric actuator is related to the tension of the string. In addition, since the current can be measured using several electronic amplifiers driving the electric actuator, measuring the current used in the electric actuator is a cost-effective method of estimating joule tension. Although the electronic amplifier for the motor cannot measure the current, the clamp-on current sensor used in the motor can be used to measure the current. The clamp on current sensor may be installed in any portion of the cable that powers the electric actuator 12. Clamp-on current sensors are Hall effect sensors that detect magnetic field strength around wires that are proportional to the original current flow. In a preferred embodiment of the present invention, the amplifier powering the electric motor has a built-in sensor that measures the current consumed by the electric motor of the actuator 12, thereby estimating the tension of the string.

FIG. 13 shows a unique auxiliary device with a clamp on current sensor 100 used to detect the current used in the actuator. Signals representing the measured values of the current transmitted by the cable 23 from the power supply of the controller 20 to the actuator 12 and the current used by the motor are transmitted to the controller via the signal cable 101. How the current measurement can be used to detect or estimate the tension of the string will be described later, but first we will describe how the tension of the found string can be used to prevent slack of the string.

Once the tension of the string is measured or estimated by the method described above, the actuator speed should be modified in accordance with the measured or estimated string tension. If the tension of the string is zero, the input to the actuator should be modified so that the speed of the actuator is zero, so that there is no extra string released. This can be done by introducing the variable K _M in the control block diagram, as shown in FIG. When the transfer function of the controller is expressed as (K), the output (e) of the controller is as follows.

14 shows that the line velocity can be represented by equation (18).

Here, K _M is a variable such that K _M = 1 when the string tension f _R is not zero. If the tension of the string is not zero, substituting K _M = 1 into equation (18) yields equation (19).

Equation (19) is similar to Eq. (6), thus saying that the action described above is still valid in three experiments. When the tension f _R of the string is detected as zero by the above-described predetermined method, (K _M ) should be changed to zero. In Eq. (18) we substitute 0 for (f _R ) and (K _M ) to get a value of zero for the line speed (v). This means that there will be no unwrapped lines and slack of the lines will be prevented if (f _R ) is detected as zero. For example, if the operator moves the end actuator downwards, the tension of the string will be non-zero, whether or not a load is connected to it. If the operator contacts the end effector with the obstruction, the weight of the end effector (and any load connected to it) will be supported by the load or obstruction and the tension of the string will be zero. At the same time the operator's operating force can be detected, which causes the rope to unwind momentarily so that the rope is no longer taut (i.e. the tension of the rope is zero) and the operator's operating force (f) is no longer dependent on the movement of the rope. Slack in the line is prevented without contributing.

Although the present invention prefers to program the system to obtain tension of the string to prevent slack, there are other ways to prevent string slack. Another method of searching for slack in a string during quasi static operation (low acceleration and deceleration movement) is to determine the operator's operating force (f) and tension (f _R ) to detect whether the end effector is supported by the string. ) At the same time. The first step is to calibrate the system before operation to obtain the tension of the string derived solely from the weight W _E of the end effector. During operation, the values of the operator's operating force f and the string tension f _R at the end operator are obtained simultaneously. Then, by subtracting the operator's operating force (f) from the tension (f _R ) of the string, the controller can separate the load (p) using equation (3). Finally, it is possible to determine whether the end effector is supported by the string by comparing the value of (p) with the stored value W _E. As long as the load load p is approximately equal to the weight W _E of the end effector, the system will know that the end effector is neither coupled to the load nor supported by the obstruction and that it is safe to unwind. If at any moment the load (p) is not equal to the weight of the actuator (W _E ) anyway, the system will know that the end actuator is supported by some obstruction, and the speed of the actuator to prevent slack of the string Will be adjusted to zero.

15A shows the change in (K _M ) as a function of (f _R ), where (K _M ) changes from 1 to 0 when the tension of the string changes from a non-zero value to zero. If the tension of the string is detected as zero, the actuator speed will be zero and the actuator will not loosen the string. This is important to ensure that the system can deviate from slack control if the operator starts moving upward relative to the end effector. However, because K _M = 0, the upward movement of the operator does not generate any string tension and terminates the slack control mode when FIG. 15A is always used to model (K _M ). This suggests that the use of the FIG. 15A plot allows the system to prevent slack but the system cannot escape slack control.

In order to solve this problem, the plot of FIG. 15B is used when the signal representing the force of the operator means upward movement, and when the signal representing the force of the operator means downward movement, FIG. Use the plot of 15a. The plot of FIG. 15B has a non-zero value of C1 for (K _M ) when the tension of the string is zero. A non-zero (K _M ) value causes the actuator speed value to be a non-zero but small value when an upward movement is initiated by the operator. This leaves the system out of slack control, so that the end effector is lifted when the operator starts moving upwards. Various functions can be used between (K _M ) values for smooth movement.

Provided that the force detection device is able to measure the tension of the string perfectly (e.g., FIGS. 8A, 7B, 8, 9 and 13), FIG. 15C is a signal representing the force of the operator on the end effector. When meant in the direction, it can be used to indicate a change in (K _M ) which is a function of the tension of the string. Smooth movement between two values of (K _M ) as a function of string tension reduces the device's sudden operation. FIG. 15D shows the change in (K _M ) as a function of the tension of the string when a signal indicative of the operator's force on the end effector implies upward movement. Note that a non-zero C1 value of (K _M ) when the string tension is zero will cause the system to deviate from slack control when a signal representing the operator's force on the end actuator implies upward movement. shall. Various mathematical functions may be used to represent the plots of FIGS. 15C and 15D. For example, equation (20) is a good candidate for mathematically representing the plots of FIGS. 15C and 15D.

Here, C ₁ is a value other than 0 when the signal representing the force of the operator on the end effector implies upward movement, but is less than 1 (unity). Equation (20) obtains the plot of FIG. 15C when C ₁ is selected as zero. C ₂ may be chosen such that the slot has an appropriate slope. If C ₂ is a large value, a larger slope is obtained for the plot of equation (20). In one embodiment, C ₁ and C ₂ were chosen to be 0.4 and 600, respectively. The change in K _M as shown in FIGS. 15A-15D can be programmed in the controller 20.

Preventing slack from detecting that the string tension is zero can be used to prevent only the string loosening without the effort of rolling up the string. The upward force signal of the operator can be acted upon by the upward winding line even if the upward force of the rope is zero when the upward signal occurs.

16 illustrates one embodiment of the present invention that can be used to provide slack protection and depalletizing from a pallet.

As shown in Fig. 16, it is shown that the end actuator is restricted to move downward without the occurrence of slack in the string when the end actuator is pushed downward by the operator. The end actuator 14 is connected to an electric actuator 12 mounted to a ceiling or overhead crane. As the shaft rotates the pulley, the rotation of the pulley causes the string 13 to raise or lower the end actuator 14 and the box 25. Two suction cups 18 are used to couple the box 25 to the end effector 14. The actuator 12 is controlled by the electronic controller 20. The computer located in the controller 20 transmits two signals through the cable 21, signals representing the operator's force from the end actuator 14 and signals representing the current consumed by the actuator 12 from the current sensor. Receive. Since the current sensor used in this embodiment is in a power amplifier (located in the controller 20) for supplying the electric actuator 12, a signal representing the current consumed by the actuator 12 is not shown in FIG. . The computer in the controller 20 sets the speed at which the pulley 11 should rotate based on two signals representing the operator's force on the end effector 14 and the tension of the string 13. The controller 20 supplies power to the actuator 12 via a cable 23. The result of the movement of the actuator 12 and the pulley 11 is sufficient to raise or lower the string 13 to an exact distance that generates sufficient mechanical force to assist the lifting or lowering operation of the operator as needed. When the worker's hand pushes the handle 16 upwards, the pulley 11 rotates to pull the string 13 upwards to lift the box 25. When the worker's hand pushes the handle 16 downward, the pulley 11 rotates so that the string 13 moves downward to lower the box 25. However, as shown in Fig. 16, while the end actuator is restricted from moving downward, the string is not loosened when the end actuator is pushed downward by the operator.

Here, we describe how the measurement of the current consumed by the actuator when the electric actuator is used in the system can be used to estimate the tension of the string. The amount of torque generated by the actuator 12 to rotate the pulley 11 and raise the load is proportional to the current used by the actuator 12. This is represented by equation (21).

Here, (T _T ) is the total torque generated by the actuator 12, (I) is the current used by the actuator 12, and (K _T ) is a proportional constant. The value of (K _T ) is usually provided by the actuator manufacturer. (K _T ) can also be measured experimentally by measuring the current consumed by the actuator for any known load on the actuator. Although Equation (21) is widely reported as the actual relationship between the torque generated by the actuator and the current consumed by the actuator, certain torques do not occur depending on the quality of the power amplifier that powers the actuator. The residual current of can be measured. The power amplifier must be adjusted to account for this redundant bias current measurement. The amount of torque T _L available for lifting the load and end actuators is equal to the difference between the total torque generated by the actuator 12 and the torque required to rotate the pulley 11 and all rotating components of the actuator. . This is represented by equation (22).

Here, T _P is the torque required to rotate the pulley 11 and all rotating components of the actuator. The torque T _P is calculated by equation (23).

here,

I _P = moment of inertia of the pulley and all rotating components of the actuator (motor and transmission) reflected on the motor shaft

B _P = coefficient of friction of the same components above

α = angular acceleration of the electric motor shaft

ω = angular speed of electric motor shaft

T _O = constant torque due to the coulomb firction of the system.

Both α and ω (angular acceleration and angular velocity of the motor shaft) can be estimated by measuring the motor shaft angle using several standard estimation techniques.

(I _P ) and (B _P ) are two parameters associated with the actuator and can be measured experimentally. (B _P ) represents the proportional constant of the torque of the motor speed during its steady state operation (ie constant actuator speed). One skilled in the art must measure the torque required to rotate the motor shaft at a constant speed. (B _P ) is the proportional constant between the steady state speed of the motor and the required torque. (I _P ) is a proportional constant of the torque with motor acceleration during movement at high accelerations. There are various ways to measure (I _P ) and (B _P ) using standard parameter estimation techniques. For example, Extended Kalman Filter is a well known method for parameter estimation and can be found in the contro sciences literature. Shankar Sastry and Marc's book "Adaptive Control" [Prentice Hall, 1989] and George Box and Gwilym Jenkins' book "Time Series Analysis" [Hgolden-Day, 1976] are excellent references for estimating models. Two simple experiments can be used to measure BP and IP.

(I _P ) can be measured by driving an actuator having a high frequency sinusoidal input torque. At high frequencies, the torque over frictional torque is rather small compared to the inertial torque due to acceleration, and (I _P ) is proportionally constant between motor acceleration and motor torque. The value of (I _P ) can be obtained by measuring the acceleration and torque of the motor shaft. By measuring (B _P ), the actuator can be driven at a constant speed. At a constant speed, the torque associated with the inertial torque due to acceleration is zero, and (B _P ) is proportionally constant between the motor speed and the motor torque. The value of (B _P ) can be obtained by measuring the motor shaft speed and torque.

(TO) is a small and constant torque due to dry friction of the actuator (especially in the transmission portion of the electric actuator). High performance, well-lubricated electric actuators have low frictional forces and (TO) are negligibly small amounts and can be measured experimentally.

Substituting (T _P ) of Equation (23) and (T _T ) of Equation (21) into Equation (22) gives the following equation for the torque required to raise the load.

(T _L ) can be calculated from Equation (24) by measuring the current of the actuator 12 and the speed and acceleration of the actuator shaft. The tension (f _R ) of the string is

Where R is the radius of the pulley 11. When the actuator has a gear head with a very large transmission ratio (a system in which reverse operation is impossible), the motor torque supporting the tension of the string is usually compared to the motor torque that accelerates (or decelerates) the rotating part of the actuator. small. In other words, the current used to provide the torque to maintain the tension of the string when high transmission ratios are used allows only a small fraction of the current consumed by the electric motor. Moreover, actuators with low transmission ratios will result in a larger range of current attempts than actuators with high transmission ratios for current readings due to changes in tension.

It should be noted that Eq. (23) shows the basic and linear form of the dynamics of the actuator. If the actuator is properly designed and lubrication is good, Equation (23) properly determines the dynamics of the system. In examples that require higher precision, the following equation (26) can be used, which is similar to equation (25) with a frictional force modeled by a non-linear relationship g (ω).

The structure of g (ω) can be estimated experimentally using standard system identification techniques. Again, the Extended Kalman Filter in parameter estimation is well known and can be found in the control scientific literature.

The slack control methods described herein are designed based on the application of a device using a suction cup. Although the attraction amplifier device is not employed in the case with a suction cup, the above-described slack control method is preferably implemented in this device. There are several states where an operator may accidentally push a portion of the load interface over several surrounding objects, including the object to be moved. The downward residual force of the operator will generate slack in the string if the end effector is prevented from moving downward. Therefore, it is always important to prevent line slack.

The investigation in FIG. 13 teaches that the change in load weight (ΔW) brings a small change in the operator's force (Δf) when (S) is a small amount. In other words, the operator would have little feel for the change in load weight if (S) is a small amount. If the tension (f _R ) of the string is measured or estimated as described above to prevent slack, the system can be further improved by using (f _R ) to more clearly produce the operator's feeling when the weight of the load changes. Here is how to achieve this improvement. Once the string tension (f _R ) of the string is known, the load (p) can be calculated from equation (3). The load load p can then be used as a feedback signal.

Where (Q) is the controller transfer function that acts on the load (p). Through this equation, (Q) can also be referred to as the feedback transfer function of the consistent force because it feeds the load back to the controller. The comparison of equations (17) and (27) shows how both the operator force and the load are used as feedback signals in equation (27). Substituting equation (27) into equation (2) and performing the same mathematical process described above yields equation (28) for line speed v.

Equation (29) shows that a predetermined change (ΔW) of the load weight and a predetermined change (Δf) of the force exerted by the operator on the end actuator will result in a change in the end actuator speed change (Δv).

The load force feedback transfer function (Q) increases the overall sensitivity of the system to the load from (S) to (S + GQ). S 'is the following when the apparent sensitivity of the article is S'.

Equation (29) can be rewritten as

Equation (31) is similar to Equation (11), but the sensitivity to load in the system is increased from (S) to (S '). Moreover, all the properties in the three experiments described above are still valid. For example, the effect of selective load feedback in Experiment 1. When the load load feedback transfer function Q is used, equation (13) can be rewritten to equation (32).

The comparison of equations (13) and (32) shows that a predetermined load weight increase is expected if the operating speed is to remain unchanged because (S ') is greater than (S), where load load feedback (Q) is used. Explain that you will increase the upward force of the worker as needed. In other words, given the increase in load weight, the operator feels more force when load load feedback is used. The use of load load feedback is optional. If (S) is large enough to give the operator a proper sense of the load load change, there is no need to perform the load load feedback. If (S) is small, the load load feedback is performed by the operator. It will improve the system's performance in the senses as it makes the senses more clear.

Here, two simple modifications of equation (27) will be described. Since the force (f) of the operator is usually small compared to the load (p), (p) in equation (27) can be replaced by (f _R ).

It is also possible to use pL (load only due to the weight and inertia of the load) if (p _L ) is readily available as shown in the example of FIG. 7B rather than using the load as feedback.

18A and 18B show a flow chart of a computer program that can be used in the controller 20. The control program first initializes all input and output hardware in the system. This includes analog-to-digital converters, digital-to-analog converters and quadrature counters in addition to all other peripherals in the controller. After calculating all the constants needed for the controller, the controller will release the friction brake on the actuator and turn on the green light on the controller indicating that the system is ready for operation. The controller then enters the main control loop to read the position of the actuator, the operator's force, the actuator's current, and the deadman switch. The controller then implements a transfer function (K) on a signal representing the operator's force. The transfer function (K) should be chosen to ensure closed loop system stability. Using the actuator position value, the controller estimates the tension of the string by equation (17) above. Using the operator's force value, the software determines whether the operator's force applied is downward (+) or upward (-). Depending on the direction of the operator force, the software calculates the K _M value in a plot similar to FIGS. 15B and 15C. Since the value of K _M is obtained from a plot similar to FIG. 15B or FIG. 15C, there will be a discontinuity in the calculation of the size of K _M. Moving to different K _M values can be done smoothly by using digital filters. Digital filters are therefore designed as high frequency components in relation to K _M. In this embodiment, a digital low pass filter has been written in software to smooth the K _M value.

The software then checks whether the Deadman switch is pressed or not. If the deadman switch is pressed, the software sends the modified value of (e) to the actuator. If the deadman switch is not pressed, the software uses the position controller to keep the actuator in its current position and engage the friction brake. This friction brake interlocks to prevent the actuator from rotating when the deadman switch is released. This friction brake adds greater rigidity to the system when the operator does not care for the device. As an additional safety feature, it is advisable to engage the friction brake whenever there is a power failure.

Many hoists use intermediary devices such as valves, push buttons, keyboards, switches or teach pendents to adjust the speed of lifting or lowering the object to be moved. In a valve controlled hoist, for example, the more the operator opens the valve, the higher the lift speed of the object. By providing the intermediary device, the operator is busy to operate the valve or switch while having no sense of how much load load has been raised because his hand is not in contact with the object. However, the operator can activate an intermediary device (eg lowering push button) to lower the load while the load is restricted to move downward. The slack prevention method described above can be used with such a universal hoist. In other words, the switches and sensors described herein (e.g., Figures 9, 10a and 10b, 11a and 11b) provide information about the tension of the string (e.g. the magnitude of the string tension or lack of the string tension). It can be used with these devices to send to. In addition, when these devices are powered up, the tension of the cord from the current measurement can also be estimated as described above.

Although specific embodiments of the invention have been shown in the accompanying drawings and described in the foregoing detailed description, the invention is not limited to the embodiments described, but all alternatives, equivalents, and modifications fall within the scope of the invention as defined by the appended claims. It should be understood that it is intended to include alignment of water and / or elements. For example, in many of the embodiments described above, the operator's operating force is used as the input to the system, but the advantages provided by the present invention, particularly load weight sensitivity and slack protection, require the use of valves or up-down switches to move loads. It is also beneficial for the hoist to use. There are also alternative ways to program the system to achieve the desired specific function, although specific equations have been published to explain system operation. The appended claims are intended to cover all such modifications and alternatives.

Claims (90)

delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete delete An end effector connected to the load to be held and lifted by the operator, a sensor in the end effector that senses the force exerted by the operator during the lift and transmits a signal to the controller indicating the force exerted by the operator, and the operator A hoist actuator for varying the speed of movement of the cord for transmitting tension from the actuator to the end effector to assist Further comprising a load load estimation device for transmitting a signal indicating the load of the load to the controller, The controller controls the actuator as a function of the estimated load load signal and a signal of the force exerted by the operator. Hoist system. The method of claim 36, The speed of the actuator is controlled as a function of the load load signal estimated by the equation v = GK (ff _up ) + GQp + S (f + p) and the signal of the force exerted by the operator, where v is the Speed, G is the actuator transfer function, K is the transfer function of the controller operating for f, f is the force exerted by the operator, f _up is constant, Q is the transfer function of the controller operating for p, p is the estimated load load And S is an actuator sensitivity transfer function. The method of claim 36, Hoist system, characterized in that the load load estimation device is a force sensor. The method of claim 38, Hoist system, characterized in that the force sensor estimates a load of load as a function of the tension of the string. The method of claim 38, Hoist system, characterized in that said force sensor estimates loads as a function of tensile stress on said hoist. The method of claim 36, And the load-bearing estimator uses the current consumed by the actuator to estimate the load-bearing. The method of claim 41, wherein And the load-bearing estimator includes a sensor for detecting a current flowing to the actuator. The method of claim 36, And the controller is programmed so that the worker always bears a part of the load while the load moves so that the worker feels the load during the lifting movement. The method of claim 36, Wherein the controller is programmed such that a change in force and load applied by the operator is functionally correlated to require a change in pulling speed or force applied in response to the change in load. The method of claim 36, And the controller responds to the detection that the tension of the string is insufficient by preventing the string from loosening in response to a downward force applied by an operator. An end actuator that is connected to the load to be held and lifted by the operator, a sensor in the end actuator that measures the force exerted by the operator and sends a signal to the controller indicating the force exerted by the operator, and assists the operator in lifting the article A method of improving the responsiveness of a hoist comprising a hoist actuator by varying the speed of movement of the string that transmits tension from the actuator to the end effector. Estimating the load of the load with the load load estimating apparatus, Transmitting a load load signal from the load load estimator to the controller, and Programming the controller to control the speed of the actuator as a function of the load load signal and a signal of the force exerted by the operator. Hoist responsiveness improvement method comprising a. 47. The method of claim 46 wherein Programming the controller to control the speed of the actuator as a function of the load load signal and the force exerted by the operator is performed according to the formula e = K (ff _up ) + Qp, where e is the output of the controller, K is f The transfer function of the controller operating for, f is the force exerted by the operator, f _up is the constant, Q is the transfer function of the controller operating for p, and p is the estimated load load. . 47. The method of claim 46 wherein And using a force sensor to estimate the load. The method of claim 48, A method for improving hoist responsiveness, comprising using a sensor to measure the tension of a string to estimate load loading. The method of claim 48, A method for improving hoist responsiveness, comprising using a sensor to measure mechanical tensile stress on the hoist to estimate load loading. 47. The method of claim 46 wherein A method for improving hoist responsiveness, comprising using a device to measure the energy consumed by the actuator when raising the load to estimate the load. 47. The method of claim 46 wherein And said actuator programs said controller to hold said load in a fixed position when a deadman switch on said end effector is inactive. An end effector connected to the load to be held and lifted by the operator, a sensor in the end effector that senses the force exerted by the operator during the lift and transmits a signal to the controller indicating the force exerted by the operator, the operation set by the controller A system for improving hoist responsiveness to an operator's input of a hoist comprising a hoist actuator having a speed, and a string transferring tension between the actuator and the end actuator to assist the operator in lifting the load. It further includes a load load estimator for measuring the force of the string generated by the lifting load, The load load estimating apparatus transmits a load load signal to the controller. The controller changes the pulling speed as a function of the load load signal in response to the force exerted by the operator. System for improving hoist responsiveness. The method of claim 53, And the load load signal includes the force of the string generated by the end effector. The method of claim 53, And said baggage load estimator measures the tension of said string. The method of claim 53, And said baggage load estimator measures mechanical tensile stress on said hoist. The method of claim 53, And the load-bearing estimator measures the energy consumed by the actuator supporting the load. The force exerted by the operator on the end effector connected to the load and connected to the hoist by the rope gives the operator a real sense of the lifting work instructed by the worker, and a signal indicative of the force exerted by the worker during the lifting operation A hoist system, executed by a controller, in which an actuator connected to the end actuator by the string is operated. The hoist system includes a load load measuring device for supplying a load load signal to the controller, The controller operates the actuator to variably perform the pulling operation indicated by the operator in response to the load load signal and the signal of the operator's force, A hoist system, characterized in that it is programmed to exert a greater operator force to pull at the same speed for a load that is heavier than necessary for lifting a light load. The method of claim 58, And the load load estimating device is a force sensor arranged to detect the load of the load. The method of claim 58, And the load load estimating apparatus estimates the load load signal from a current required by the actuator to support the end actuator and the load connected to the end actuator. The method of claim 58, For a given operator's downward force, an increase in load weight increases the rate of descent of the load. The method of claim 58, A hoist system, characterized in that an increase in weight without changing the speed of the load requires more force of the operator to raise the load and less force of the operator to lower the load. Generating a load load signal indicative of the force generated by the load being lifted, and Using the load load signal to drive the hoist to increase the force exerted by the operator to lift the heavy load at the same speed as the light load Including; The hoist operator's manpower input and pulling work signal is made by the hoist to realize the weight of the load on which the worker is to be lifted. Hoist control method. The method of claim 63, wherein And obtaining the load load signal from the current required by the hoist to support the load. The method of claim 63, wherein The load control signal is obtained from a force sensor. The method of claim 63, wherein And reducing the force of the operator required to lower the heavy load at the same speed as the light load. The method of claim 63, wherein An increase in the weight of a load without changing the force exerted by an operator reduces the speed of rise of the load and increases the rate of descent of the load. The method of claim 63, wherein A method of controlling a hoist, characterized in that an increase in the weight of a load without a change in the moving speed of the load requires more force of the operator to raise the load and less force of the operator to lower the load. A load load estimating device generating a signal as a load load function supported by the hoist, and The controller driving the actuator of the hoist to change the hoist assistance for the operator such that as the load of the load increases the force of a given operator reduces the rate of rise of the load and increases the rate of descent of the load. Hoist control method. The method of claim 69, wherein And the load load is estimated from the current required for the actuator to support the load. The method of claim 69, wherein And said load is estimated from energy consumed by said actuator to support said load. The method of claim 69, wherein And said load is estimated from a force sensor. A hoist system for controlling a pulling speed of a load in response to a worker input, including a controller receiving a load load signal and a signal indicative of the force of the worker. The controller is programmed to control the rate of rise as a function of the two signals, and also the increase in load weight and the change in force of the operator is (1 + SE + GQE) Δv = (GK + S) Δf + (S + GQ Is further programmed to cause a speed change of the load according to the formula Where S is the actuator sensitivity transfer function, E is the kinetic energy of the load, G is the actuator transfer function, Q is the controller transfer function that acts on p, Δv is the change in end actuator velocity, Δf is the change in operator force, and K is The transfer function of the controller operating for f, and ΔW represents the change in load weight Hoist system. The method of claim 73, The controller has a higher upward force and smaller than necessary to move the light load at the same speed with respect to the heavy load, as evidenced by the formula 0 = (GK + S) Δf + (S + GQ) (ΔW). The operator is programmed to apply downward force, where G is the actuator transfer function, K is the transfer function of the controller operating for f, S is the actuator sensitivity transfer function, Δf is the change in force applied by the operator, and Q is p A controller transfer function operating for, and ΔW, wherein the hoist system is characterized by a change in load weight. The method of claim 73, The controller is programmed such that for any given weight any increase in the speed of movement of the load requires additional operator force according to the formula (1 + SE + GQE) Δv = (GK + S) Δf, where S Is the actuator sensitivity transfer function, E is the kinetic energy of the load, G is the actuator transfer function, Q is the controller transfer function operating for p, Δv is the change in end effector velocity, and K is the transfer function of the controller operating for f And Δf represents a change in operator force. The method of claim 73, The change in load weight for a given operator force results in a corresponding speed change of the end effector according to the formula (1 + SE + GQE) Δv = (S + GQ) (ΔW), where S is the actuator sensitivity transfer function Where E is the kinetic energy of the load, G is the actuator transfer function, Q is the controller transfer function operating for p, Δv is the change in the end effector speed, and ΔW is the change in the load weight. The method of claim 73, A hoist system comprising a load load estimator formed as a force sensor. The method of claim 73, And a load load estimator using the current consumed by the actuator to estimate the load. A control system for a hoist having a controller of an actuator and a hoist string extending from the actuator to an end effector that can be connected to a load. A detector which generates a signal to the controller whenever there is no tension in the string, The controller prevents the actuator from unwinding in response to an operator's downward movement input whenever a signal is generated indicating the absence of tension in the string. Control system for hoist. The method of claim 79, Hoist control system, characterized in that the detector is a switch connected to the cord. The method of claim 79, And said detector measures a decrease in tension of the cord by detecting a load of load and an operator input. The method of claim 79, And the detector measures a decrease in tension of the string from the current required by the actuator supporting the string. The method of claim 79, And said detector is capable of generating a binary signal having a first state when the tension of the string is zero and a second state when the tension of the string is not zero. The method of claim 79, The detector is capable of generating a binary signal having a first state in which the end actuator is restricted to move downward and a second state in which the end actuator is not restricted to move downward. Control system. The method of claim 79, And the detector includes a switch that can move to one position if slack occurs in the string and to a different position if slack does not occur in the string. The method of claim 79, Wherein the controller generates a rising speed command signal to generate a non-zero tension in the string in response to an upward movement input of the operator whenever the signal from the detector indicates that the string is absent in the hoist. Control system. A method of preventing slack from occurring in a hoist string used in a hoist system, Detecting the tension member of the string; Generating a signal to the controller of the actuator of the hoist each time there is no tension in the string, and Programming the controller to prevent the actuator from releasing the string in response to an operator's downward movement input when the tension of the string is reduced. How to include. 88. The method of claim 87 wherein Detecting the tension of the string with a switch connected to the string. 88. The method of claim 87 wherein Detecting the tension of the string from the load and operator input. 88. The method of claim 87 wherein Detecting the tension of the string from the energy consumed by the actuator supporting the string.