Apparatus, gripper apparatus and soft robot gripper system - Patents.com

Abstract

To provide a soft robotic gripper with improved operating performance level quantity.SOLUTION: A soft gripper includes tentacles, each tentacle includes an upper member and a lower member connected by a connector. Each member includes guide discs, and each guide disc includes a ring with through holes, and a spacer having the through holes is located in a donut hole of the ring, the through holes collectively define cable pathways. The connector includes a center thru-hole and transfer channels. Cables have proximal ends attached to actuators and extend through apertures of a baseplate located at the proximal end of the lower member. A set of lower cables extend through the through holes of the lower ring to couple to a distal lower guide disc. A set of upper cables passes through the through holes of the lower spacer, extends through the transfer channels to the through holes of the upper ring to couple to the distal upper guide ring, and an end cap is attached to a distal end of the upper member.SELECTED DRAWING: Figure 1A

Description

[Technical field]

[0001] Field The present disclosure relates to a soft robotic tentacle system, and more particularly to a tentacle that is independently actuated by a transmission system in which an upper cable of the transmission system extends through a lower spacer passage, through a transmission channel to an upper ring passage, uniquely providing minimal cable transmission friction, and the tentacle includes a semi-full or full actuation mechanism to move objects within a workspace. [Background technology]

[0002] background Conventional soft robotic grippers and / or end point effectors can be classified as pneumatic grippers, hydraulic grippers, electromagnetic motor grippers, and suction and vacuum grippers. Each conventional soft robotic gripper exhibits several advantages but also comes with numerous problems. For example, some of the problems that conventional soft robotic grippers have are too large in overall weight and design, low durability, wear and tear over time, slow gripping in terms of the time the gripper holds an object, and limited or sometimes no control for fine manipulation. Some other problems include high maintenance requirements that limit the application areas.

[0003] Regarding conventional hydraulic grippers, these devices include hydraulic cylinders that slide or pivot the fingers together and apart. Some aspects of conventional hydraulic grippers include a strong gripping force (over 10,000 Nt) that is powerful to bend and clamp thin steel sheets such as car body panels. However, conventional hydraulic grippers require a hydraulic power source. In addition, some issues with conventional hydraulic grippers are that they have a slow take-up time to operate the gripper retraction and their total weight is very heavy. Other issues include an inflexible 2000-6000 PSI hose. There are also environmental concerns with the use of conventional hydraulic grippers, as typical leakage creates a virtual EPA problem when cleaning after a leakage.

[0004] With regard to conventional electromagnetic motor grippers, these devices typically include an electromagnetic motor (servo, stepper, or similar motor) that actuates fingers arranged via a gear train. Conventional electromagnetic motor grippers require electricity and a control CPU to operate. However, some problems with conventional electromagnetic motor grippers are that the geared units have a very slow uptake time and a weak grip compared to a conventional air-driven unit of the same weight. Other problems include a very poor grip strength to gripper weight ratio and a heavy overall weight. Further problems include limited overall strength due to heat dissipation to the motor.

[0005] Regarding conventional suction and vacuum grippers, these devices include a vacuum pump or vacuum venturi that creates a low pressure area, which sucks the gripped object to a nozzle, which then holds the object. Conventional suction and vacuum grippers require electricity or compressed air to generate the vacuum. Some problems with conventional suction and vacuum grippers are that the acquisition time is very slow (more than 3 seconds) and the gripper strength is weak. Other problems include that they do not "position" and the pose of the gripped object is not controlled (i.e. the position in the gripper is unknown), and another problem is that the vacuum area does not allow access to the periphery of the gripped object.

[0006] With respect to conventional biomimetic / anthropomorphic grippers, these devices include grippers that attempt to be analogues of animal or human organs of manipulation. For example, the "tentacle" gripper attempts to combine the foot movements of an octopus with the vacuum suction cups of multiple cephalopods, using pneumatically powered arms and grippers. Summary of the Invention [Problem to be solved by the invention]

[0007] There is an urgent need to develop soft robotic grippers with improved levels of operational performance to meet the growing variety of work applications in the gripper industry. For example, some of the improved gripper operational performances are reduced total gripper weight, increased gripping force, increased gripping speed, and increased gripper control, which in effect results in dexterity compared to the operational performance levels offered by conventional grippers. There is also a need for soft robotic gripper technology to provide grippers with increased degrees of freedom of movement (DoF) within the gripper operational travel space (i.e., similar to the movement of a human finger when writing in cursive), increased positioning (i.e., biologically similar to proprioception), and high levels of force feedback (biologically similar to muscle movement). [Means for solving the problem]

[0008] overview The present disclosure relates to a soft robotic tentacle system, and more particularly to a tentacle that is independently actuated by a transmission system in which an upper cable of the transmission system extends through a lower spacer passage, through a transmission channel to an upper ring passage, uniquely providing minimal cable transmission friction, and the tentacle includes a semi-full or full actuation mechanism to move objects within a workspace.

[0009] Some embodiments of the present disclosure include a soft robotic gripper with a mechanical structure design that provides 12 degrees of freedom (DoF) of coordinated tentacle gripping motion to grip / grasp objects with safe, controllable, precise and dexterous manipulation. At least one realization emerging from experimentation is that multi-tentacle systems must meet today's gripper performance levels for industrial applications by being lightweight in design and durable enough to withstand long-term operational wear and tear.

[0010] Various types of experimental conventional gripper designs were constructed and tested, including pneumatic grippers, hydraulic grippers, electromagnetic motor grippers, etc., for example, in industrial applications. It was found that these test grippers lacked performance levels that meet many of today's gripper applications. Several problems with the grippers were identified and needed to be overcome, including, among others, (a) the grippers being too heavy, (b) low resistance to long-term operational wear and tear, (c) slow gripping times in terms of the length of time the gripper grips an object, and (d) high maintenance costs. For this reason, further testing was not pursued with these types of experimental conventional gripper configurations. More experimental testing was then conducted on new test grippers designed to overcome the above problems and provide other advantages.

[0011] Some of the embodiments of the present disclosure include grippers or tentacles that are designed such that each tentacle is composed of a lower elastic tube or member and an upper elastic tube or member connected by an elastic connector. The lower and upper elastic tubes may be identical in length and diameter, and a guide disk is attached inside each tube along its length to maintain the spacing between the guide disks. The guide disks may be identical in construction, and each guide disk includes an outer ring with a through hole and a spacer disposed in a donut hole of the ring, also with a through hole. The ring and spacer through holes together define the cable path. Fillets are configured at the inlet and outlet openings of the transmission channel, the lower and upper connector center through holes, the lower and upper ring through holes, the lower and upper spacer through holes, and the through holes of the guide ring. Furthermore, one or more of the guide disks may include convex bottom and top surfaces and rounded peripheries or convex edges. As another aspect of the guide rings, some of the guide rings may be constructed such that the stiffness of the material forming the guide rings is different than the stiffness of the material forming the spacers.

[0012] It is contemplated that the length and diameter of the lower elastic tube may differ from the length and diameter of the upper elastic tube (see Figures 1A-1E), and that the thickness and diameter of the guide disc may vary (see Figures 2D-2I), all depending on the operational and performance requirements of the user.

[0013] The elastic connector is configured with a central through hole that runs the entire length of the connector, and is designed with a transmission channel in the central portion of the connector (see FIG. 1F). Each transmission channel may include a geometric feature having an entrance opening with a cable entrance path tangent to the inner surface of the central through hole and an exit opening with a cable exit path parallel to the central axis of the connector. The entrance and exit openings of the geometric feature are thus generally aligned with the upper ring through hole of the upper elastic tube and the lower spacer through hole of the lower elastic tube.

[0014] The above geometric features (see Figs. 2A-2C) are based on a mathematical formula for optimizing the transmission channel configuration with a central through hole, resulting in a minimum amount of side load on the cable resulting from the cable's operational movement, and simultaneously a minimum amount of transmission friction and cable bending fatigue during the operation of the tentacle gripper. During testing to configure the connector of the present disclosure, at least one realization was obtained that the spacer through hole of the lower elastic tube needs to be designed to align with the inlet opening of the transmission channel of the connector's geometric features, and the outlet opening of the transmission channel needs to be aligned with the through hole of the upper ring. As a result, based on further connector testing of the cable inlet path, the cable inlet path is configured to extend from the bottom surface to the top surface of the elastic connector at an acute angle, and the cable outlet path is configured to extend from the top surface to the bottom surface at an acute angle.

[0015] To move the tentacle, a cable is driven by a motor or actuator. The cable has a proximal end coupled to the motor or actuator external to the lower and upper elastic tubes or members and extends from the actuator through an opening in the controllable palm base plate located at the proximal end of the lower elastic tube. A set of lower cables extends through a through hole in the lower ring of the lower elastic tube and couples to the distal lower guide disk. A set of upper cables extends through a through hole in the lower spacer of the lower guide ring, through a transmission channel in the connector, to a through hole in the upper ring and couples to the distal upper guide ring. An end cap is attached to the distal end of the upper member.

[0016] The disclosed innovative tentacle design with 12 degrees of freedom (DoF) provides a wide variety of gripping styles resulting in a high ability to grip objects of various sizes, weights, and shapes. Some specifications associated with the disclosed gripper system include (a) gripping objects with outer diameters of about 40-95 mm or 35-75 mm with a gap of about 63.5 mm between opposing fingers, (b) cycle times of 88 msec to close and 166 msec to open, (c) gripping force of about 2.5 lb per finger, and (d) most embodiments meet the requirements of FDA-CFR Chapter 21 for food contact materials, are rated IP67 ingress protection, and operate at decibels below 85 dB. The above specific tentacle gripper performance specifications provide substantially higher performance levels that exceed the expectations of modern gripper industry performance as compared to the levels of conventional test grippers in some of the modern gripper industry applications of gripper performance specifications testing.

[0017] For example, the new tentacle design can be combined with other tentacles to grip irregularly shaped objects, including inorganic irregularly shaped objects (rocks, man-made objects) and organic irregularly shaped objects (raw raw fruits, processed foods, baked foods, non-baked foods, animal products (meat, eggs, cheese)). The various types of shaped objects that the innovative multiple tentacles can carry include, by way of non-limiting example, three-dimensional (3D) shapes, including shapes that are the same or similar to (a) polyhedrons, which are shapes with flat sides; (b) ellipsoids, which are ovoids; (c) spherical objects; (d) cylinders; and (e) cones. Shape can be understood to be the shape of the outer boundary, contour, or exterior surface of an object or objects, as opposed to other characteristics such as color, texture, or material type. Most of the conventional test grippers were unable to grip the irregularly shaped objects and / or, if they were able to grip the objects, damaged the objects or frequently dropped the objects.

[0018] The embodiments of the present disclosure are designed to grasp and move organic shapes that are often found in nature. For example, these graspable organic shapes may be more fluid and less symmetrical objects such as leaf shapes, rock shapes, and cloud shapes. Other objects grasped by the innovative tentacle may include abstract shapes including cubes, cylinders, triangles, and organic shapes such as rocks, fruits, sponges, etc., which may be represented in generalized shape methods such as point clouds. Some of the various features of the grasping capabilities of the innovative tentacle include the level of grasping non-rigid objects such as bendable and deformable objects, which may include (a) small objects, i.e. bags with bolts, grommets, etc., (b) packaged perishable goods, and (c) fragile goods such as sushi, baked goods, or other fragile goods, as compared to the level of multipurpose gripping actions associated with deformable objects by existing conventional soft robotic technologies. Experiments with this innovative tentacle have demonstrated a higher level of precision and dexterity in completing human-robot interaction tasks in an industrial manufacturing environment, enhancing performance levels when working alongside unprotected, unskilled humans, i.e., demonstrating non-injury industrial applications with humans.

[0019] Later, after some experimentation, it was found that such innovative multiple tentacles with fine manipulation can be used with motors such as radio controlled (RC) servos and cables with a bending range of + / -120° per tentacle section that can be operated via a centralized control system. Some embodiment RC servo and tension cable configurations have axial lift strengths of over 100N (i.e., lift strengths greater than 10 kg). At these axial lift strengths, grip strength levels translate into a higher tentacle grip strength to weight ratio compared to tentacles with a lighter total weight, i.e., the tentacle configuration is lighter compared to a conventional gripper of the same size. The higher tentacle grip strength to weight ratio results in a higher tentacle grip strength capacity, allowing heavier objects to be moved in less time compared to a conventional gripper of the same size with a lower tentacle grip strength to weight ratio.

[0020] After numerous long test runs of the tentacle operating in various gripper industrial applications, it was observed that the tentacle did not have RC motor heat dissipation issues. In similar long test runs of experimental conventional gripper systems, RC motor heat dissipation issues were observed, resulting in limited total grip strength.

[0021] At least one surprising attribute observed after an extended period of operational testing was the reduced gripping time to move an object from a start position to a final position compared to the gripping time required by a conventional gripper to grip the same object and move it from the same start position to the same final position. It was also observed that the RC servo motor and tension cable took less than one second to actuate the force compared to a conventional electromagnetic motor gripper. At least one reason the conventional gripper was so slow was believed to be due to the intake time of the gearing. Other problem areas observed included the weak grip strength of the conventional electromagnetic motor gripper, which may have been a result of the heat dissipation of the RC motor as mentioned in the previous test experiments.

[0022] Another observation from the test human-robot collaborative task completion application was that human workers reported that the tentacles were so deft in their operation that an unprotected, unskilled human felt confident while completing a human-robot interaction task in an industrial manufacturing environment and had little concern that the tentacles would injure him. In contrast, when the same human workers engaged in the same test human-robot collaborative task completion application using conventional grippers, most of the human workers reported that they did not feel safe and thought that they would eventually be injured by the conventional grippers, in part because the grippers lacked semi-perfect or complete actuation mechanisms and were not deft in their operation. These types of conventional grippers, which have been applied in the past to the same industrial applications, were found to be prone to injury and undermined human safety and human worker confidence.

[0023] To solve these conventional safety problems, embodiments of the present disclosure include tentacles constructed of elastic material to avoid pinch points that would cause harm to humans. The preventative measures for these embodiments include the previously mentioned geometry of the elastic connector connecting the lower elastic tube to the upper elastic tube. This geometry minimizes the amount of cable side load that occurs from the cable moving during operation while simultaneously minimizing the amount of transmission friction and cable bending fatigue during operation of the tentacle gripper. However, just as importantly, this geometry aids in highly dexterous manipulation of the tentacle and the distal articulation sensors, i.e. distributed shape or linear displacement sensors, along with the embedded tip sensor at the tip, provide a level of tentacle control that results in a nearly 100% injury free working environment for the human operator. "Human-robot collaboration is a key element in the development of the factory of the future, a space where humans and robots can work and perform tasks together. Safety is one of the most important features of this human-robot collaboration model," states Maurtua et al. in the research paper "Human-robot collaboration in industrial applications: Safety, interaction and trust" (published July 7, 2017: https: / / doi.org / 10.1177 / 1729881417716010). Best managers at modern industrial manufacturers are working with researchers to realize robot grippers that do not cause such injuries.

[0024] Another advantage of the disclosed embodiments is that the robot arm mass loading can be reduced by inserting a cable extension between the base plate of the multi-tentacle gripper and a servo motor package located and supported externally to the robot through a flexible sheathed cable, such as a bicycle brake cable. The flexible sheathed cable does not have to bend to flex the tentacles, so the lack of flexibility in the flexible sheath does not affect the dexterity of the multi-tentacle gripper. By locating the mass of the servo motor package away from the robot arm, the mass impact of the gripper is very low in terms of its force and velocity, since the motive forces for each degree of freedom are not carried by the robot, but are remote with the forces and force feedback carried by the flexible sheathed cable.

[0025] Yet another advantage of the disclosed embodiments is that the tentacles do not need to contain electronics, in fact there is no requirement for metals or conductors at all. Grippers according to the present disclosure could be made purely of polyurethane, synthetic rubber, or other elastic plastics, with nylon, aramid (Kevlar) or ultra-high molecular weight polyethylene (UHMWPE, UHMW), (Dyneema™ or Spectra™) cables and Delrin™, nylon, PEEK, Ultem™ (polyetherimide) or fiberglass reinforced epoxy spacers and rings, and with anti-friction bushings (made of nylon, Delrin™, Teflon™) for the spacers and rings. Thus, the disclosed embodiments could be used in environments with extremely high magnetic or electric field strengths, and in high RF environments. Furthermore, since no electronics are required in the tentacles, the grippers could be used in areas of strong ionizing radiation that destroys semiconductors. An additional element of the present disclosure can be a simple calibration station consisting of a precise, preferably hardened, block about the size of the distal phalanx, equipped with an axial force gauge, rigidly positioned at a known position in the system's frame of reference, and optionally having a video camera.

[0026] According to an embodiment of the present disclosure, the device comprises a tentacle, each tentacle including a lower elastic member and an upper elastic member connected by an elastic connector. Each elastic member includes a guide disk, each guide disk including a ring with a through hole, a spacer with a through hole located in a donut hole of the ring, and the through holes of the ring and the spacer together define a cable path. The elastic connector includes a central through hole and a transmission channel in a central portion. A cable has a proximal end coupled to an actuator outside the lower elastic member and extends from the actuator through an opening in the controllable palm base plate located at the proximal end of the lower elastic member. A set of lower cables extends through the through holes of the lower ring and couples to the distal lower guide disk. A set of upper cables extends through the through holes of the lower spacer and through the transmission channel to the through holes of the upper ring and couples to the distal upper guide ring, and an end cap is attached to the distal end of the upper member.

[0027] According to another embodiment of the present disclosure, a gripper device includes a tentacle, each tentacle including a lower member and an upper member connected by a connector. Each member has a guide disk that is mounted along the member to maintain the spacing between the guide disks. Each guide disk includes a ring with a through hole, a spacer with a through hole located in a donut hole of the ring, and the through holes of the ring and the spacer together define a cable path. The connector includes a central through hole and a central transmission channel, each transmission channel includes a geometric feature having an entrance opening with a cable entrance path tangent to an inner surface of the central through hole and an exit opening with a cable exit path parallel to a central axis of the connector. A cable has a proximal end coupled to an actuator that is external to the lower member and extends from the actuator through an opening in a controllable palm base plate located at the proximal end of the lower member. A set of lower cables extend through the through holes in the lower ring and couple to a distal lower guide disk. A set of upper cables extend through the lower spacer threads of the lower guide ring, through the transmission channels to the upper ring threads and connect to the distal upper guide ring. An end cap is attached to the distal end of the upper member.

[0028] According to another embodiment of the present disclosure, a soft robotic gripper system includes a tentacle, each tentacle including a lower member and an upper member connected by a connector. Each member has a guide disk that is mounted along the member to maintain spacing between the guide disks. Each guide disk includes a ring with a through hole, a spacer with a through hole located in a donut hole of the ring, and the through holes of the ring and spacer together define a cable path. The connector includes a central through hole and a central transmission channel, each transmission channel includes a geometric feature with an entrance opening with a cable entrance path tangent to an inner surface of the central through hole and an exit opening with a cable exit path parallel to a central axis of the connector. A cable has a proximal end coupled to an actuator external to the lower member and extends from the actuator through an opening in a controllable palm base plate located at the proximal end of the lower member. A set of lower cables extend through the through holes in the lower ring and couple to a distal lower guide disk. A set of upper cables extend through the through holes of the lower spacer of the lower guide ring and through the transmission channel to the through holes of the upper ring to couple to the distal upper guide ring, and an end cap is attached to the distal end of the upper member. A centralized control system is configured to receive the tentacle data and the sensor data from the transceiver. The sensor data is used to obtain a corresponding stored object configuration by comparing it with the stored object configuration. The stored commands are compared to obtain a set of stored commands corresponding to the stored object configuration. The received tentacle data is used to obtain a corresponding set of stored tentacle actions by comparing it with the stored object configuration and associated stored tentacle actions. The set of stored commands is compared to obtain a first set of corresponding commands, and if the tentacle actions from the tentacle data do not correspond to the stored object configuration, a second set of commands is selected. A set of control signals is generated to cause a motor for each of the tentacles to provide a set of tensions to the transmission system to each tentacle according to an associated set of selected commands to move the object.

[0029] According to another embodiment of the present disclosure, a soft robotic gripper system includes a tentacle, each tentacle including a lower member and an upper member connected by a connector. Each member has a guide disk that is mounted along the member to maintain spacing between the guide disks. Each guide disk includes a ring with a through hole, a spacer with a through hole located in a donut hole of the ring, and the through holes of the ring and spacer together define a cable path. The connector includes a central through hole and a central transmission channel, each transmission channel includes a geometric feature with an entrance opening with a cable entrance path tangent to an inner surface of the central through hole and an exit opening with a cable exit path parallel to a central axis of the connector. A cable has a proximal end coupled to an actuator external to the lower member and extends from the actuator through an opening in a controllable palm base plate located at the proximal end of the lower member. A set of lower cables extend through the through holes in the lower ring and couple to a distal lower guide disk. A set of upper cables extend through the through holes of the lower spacer of the lower guide ring and through the transmission channel to the through holes of the upper ring to couple to the distal upper guide ring, and an end cap is attached to the distal end of the upper member. A centralized control system is configured to receive the tentacle data and the sensor data from the transceiver. The sensor data is used to obtain a corresponding stored object configuration by comparing it with the stored object configuration. The stored commands are used to obtain a set of stored commands corresponding to the stored object configuration. The received tentacle data is used to obtain a corresponding set of stored tentacle actions by comparing it with the stored object configuration and associated stored tentacle actions. The set of stored commands is used to obtain a first set of corresponding commands, and if the tentacle actions from the tentacle data do not correspond to the stored object configuration, a second set of commands is selected. The selected set of commands includes a series of predetermined torques, which are applied to one or more joints or flexion joints and a plurality of transmission systems of each of the tentacles, along with corresponding tensions.Each transmission system has a first end connected to a motor and a second end connected to a joint or flex joint. To move an object, a motor for each one of the tentacles generates a set of control signals that cause the motor to provide a set of tensions to the transmission system to each tentacle according to an associated set of selected commands. The set of selected commands includes a set of predetermined torques, the torques along with corresponding tensions are provided to one or more joints or flex joints to provide the transmission systems of each one of the tentacles. The transmission systems have a first end connected to the motor and a second end connected to the joint or flex joint.

[0030] The presently disclosed embodiments will be further described with reference to the accompanying drawings, in which the drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the presently disclosed embodiments. [Brief description of the drawings]

[0031] [Figure 1A] FIG. 13 is a photograph showing a tentacle of a soft robotic gripper according to an embodiment of the present disclosure. [Figure 1B] FIG. 1B is a schematic diagram illustrating a tentacle of the soft robotic gripper of FIG. 1A according to some embodiments of the present disclosure. [Figure 1C] FIG. 1B is a schematic diagram illustrating some components of the soft robotic gripper of FIG. 1A according to some embodiments of the present disclosure. [Figure 1D] 1C is a schematic diagram illustrating a cable configuration of the rings and spacers of the lower tube of the soft robot gripper of FIG. 1A and FIG. 1B according to some embodiments of the present disclosure. [Figure 1E] 1C is a schematic diagram illustrating a cable configuration of the rings and spacers of the upper tube of the soft robotic gripper of FIG. 1A and FIG. 1B according to some embodiments of the present disclosure. [Figure 1F]FIG. 1C is a schematic diagram showing a cable configuration according to some embodiments of the present disclosure, the cable configuration extending from the controllable palm base plate in FIG. 1B toward the tip in FIG. 1B first enters the lower tube spacer of the distal guide disk (C1-C4), exits the lower tube spacer passage, enters the tube connector via the central through hole (D1-D4), passes through the transmission channel of the tube connector, exits the transmission channel into the upper tube ring passage (E1-E4), and exits the upper tube ring passage (F1-F4). [Figure 2A] FIG. 1 is a schematic diagram illustrating optimization of elastic connectors to minimize cable friction and cable material fatigue according to an embodiment of the present disclosure, showing a side view of the cable path. [Figure 2B] FIG. 2B is a schematic diagram illustrating optimization of an elastic connector to minimize cable friction and cable material fatigue according to an embodiment of the present disclosure. FIG. 2C shows a perspective view of FIG. 2A. [Figure 2C] FIG. 2B is a schematic diagram illustrating optimization of an elastic connector to minimize cable friction and cable material fatigue according to an embodiment of the present disclosure, showing a top view of FIG. 2A. [Figure 2D] FIG. 13 is a schematic diagram showing a top view of a guide disk having a passage with rounded or convex edges and a central through hole to minimize friction and wear on the elastic tube, according to some embodiments of the present disclosure. [Figure 2E] FIG. 13 is a schematic diagram illustrating a perspective view of a guide disk having a passageway and a through center hole with fillets to minimize cable friction and material fatigue during cable movement and rounded peripheral or convex edges to minimize friction and wear on the elastic tube in accordance with some embodiments of the present disclosure. [Figure 2F] FIG. 13 is a schematic diagram showing a side view of a guide disk having a passage with rounded peripheral or convex edges and a central through hole to minimize friction and wear on the elastic tube, according to some embodiments of the present disclosure. [Figure 2G]FIG. 13 is a schematic diagram showing a side view of a guide disk having a passage with rounded peripheral or convex edges and a central through hole to minimize friction and wear on the elastic tube, according to some embodiments of the present disclosure. [Figure 2H] FIG. 13 is a schematic diagram showing a side view of a guide disk having a passage with rounded peripheral or convex edges and a central through hole to minimize friction and wear on the elastic tube, according to some embodiments of the present disclosure. [Figure 2I] 1A-1C are schematic diagrams illustrating side views of various sized guide disks with passages and through-holes and rounded peripheral or convex edges positioned within an elastic tube according to some embodiments of the present disclosure. [Figure 3A] FIG. 1 is a schematic diagram illustrating a soft robotic tentacle gripper system according to some embodiments of the present disclosure. [Figure 3B] FIG. 13 is a schematic diagram illustrating a palm baseplate configuration in which each tentacle of a plurality of tentacles can move independently from the other remaining tentacles in an X direction along an X axis and a Y direction along a Y axis, according to some embodiments of the present disclosure. [Figure 3C] FIG. 13 is a schematic diagram illustrating another configuration of a palmar baseplate according to some embodiments of the present disclosure, where the outer edge of the palmar baseplate can be adjustable either inward or outward along the horizontal plane B (HP-B) or along the extent of the vertical plane A (VP-A) of the palmar baseplate. [Figure 3D] FIG. 13 is a schematic diagram illustrating that the palm base plate is adjustable along the Y axis so that it can be tilted towards or away from the tentacles, and is adjustable left and right along the X axis, according to some embodiments of the present disclosure. [Figure 3E] FIG. 1 is a schematic diagram illustrating a side view of a detachable partial grab device according to some embodiments of the present disclosure. [Figure 3F] FIG. 1 is a schematic diagram illustrating a side view of a detachable partial glove according to some embodiments of the present disclosure. [Figure 3G]FIG. 3B is a schematic diagram illustrating some components of a centralized controller system of the soft robotic gripper system of FIG. 3A according to some embodiments of the present disclosure. [Figure 4] FIG. 13 is a schematic diagram illustrating an embodiment of a robot assembly including incorporating a soft robotic gripper system with an elastomeric sensor on a palm base plate according to some embodiments of the present disclosure. [Figure 5A] FIG. 13 is a photographic illustration of a type of multi-tentacled grip or grasp on a molded object that may be used in industrial settings, such as manufacturing for grouping (MFG) packaging applications for small parts, or supply chain applications for bin picking, item sorting, according to some embodiments of the present disclosure. [Figure 5B] FIG. 13 is a photographic illustration of a type of multi-tentacled grip or grasp on a molded object that may be used in industrial settings, such as manufacturing for grouping (MFG) packaging applications for small parts, or supply chain applications for bin picking, item sorting, according to some embodiments of the present disclosure. [Figure 5C] FIG. 13 is a photographic illustration of a type of multi-tentacled grip or grasp on a molded object that may be used in industrial settings, such as manufacturing for grouping (MFG) packaging applications for small parts, or supply chain applications for bin picking, item sorting, according to some embodiments of the present disclosure. [Figure 5D] FIG. 13 is a photographic illustration of a type of multi-tentacled grip or grasp on a molded object that may be used in industrial settings, such as manufacturing for grouping (MFG) packaging applications for small parts, or supply chain applications for bin picking, item sorting, according to some embodiments of the present disclosure. [Figure 6A] FIG. 16 is a photographic diagram showing another type of multiple tentacle grip or grasp for a heavy object, such as a slug hammer 1610, according to some embodiments of the present disclosure. [Figure 6B]FIG. 1 is a schematic diagram showing a table of grip strength test results for a test gripper having three tentacles and two sections per tentacle, fully actuated (12 DoF) per tentacle section, + / -120° bend, and a minimum internal bend radius of 30 mm; the grip strength tested includes grip load capacity, initial and maximum current draw, and failure modes determined for several grips, according to some embodiments of the present disclosure. [Figure 7] FIG. 1 is a schematic diagram illustrating a robotic system including the integration of a soft robotic gripper system with an elastomeric sensor on a palm base plate and a human worker working together as a team to perform tasks on an assembly line of a discrete manufacturing process according to some embodiments of the present disclosure. [Figure 8A] FIG. 13 illustrates an embodiment of a tentacle attached to a "teach glove" worn by a human being that is used to create a set of motion instructions specific to each pose of an object to move the object, which can be stored in a motion instruction database, and each object can include multiple poses, in accordance with some embodiments of the present disclosure. [Figure 8B] FIG. 8B is a block diagram illustrating some method steps for creating a set of motion instructions for each pose of an object using the human-worn “teach glove” of FIG. 8A , each pose having an associated set of motion instructions, according to some embodiments of the present disclosure. [Figure 9] FIG. 13 is a schematic diagram illustrating an alternative centralized controller system that can be used to perform some methods associated with the soft robotic gripper system according to some embodiments of the present disclosure, or that can be combined with the centralized control system of FIG. 10 and / or the centralized controller system of FIG. 3E. [Figure 10]FIG. 13 is a schematic diagram illustrating an alternative centralized controller system that can be used to perform some methods associated with the soft robotic gripper system according to some embodiments of the present disclosure, or that can be combined with the centralized control system of FIG. 9 and / or the centralized controller system of FIG. 3E.

[0032] While the above drawings depict embodiments disclosed herein, other embodiments are contemplated as described herein. The present disclosure presents illustrative embodiments by way of representation and not by way of limitation. Those skilled in the art can envision numerous other modifications and embodiments that fall within the scope and spirit of the principles of the embodiments disclosed herein. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0033] Detailed Description The present disclosure relates to robotic manipulators, and more specifically to soft robotic grippers having independently actuated multiple tentacle grippers coordinated by centralized control with multiple tentacles having semi-perfect or full actuation mechanisms for moving objects.

[0034] 1A is a photograph showing multiple tentacles 111 of a soft robotic gripper according to an embodiment of the present disclosure. Each tentacle 111 includes a lower tube 118, an upper tube 119, and a tip 116. It is intended that each tentacle 111 can have an individual length, width (radius), or both to provide a unique grip type if a user requires such a tentacle configuration for an application.

[0035] It is contemplated that one or more portions of the tentacle 111, or the entire outer surface OS, may be constructed to include at least one low friction material, such as a polymer, to reduce the level of friction coefficient between the outer surface OS of the tentacle 111 and the object being gripped, thereby reducing the maximum frictional grip strength of the gripper. Subsequent experimentation has taught that one or more portions of the outer surface OS of the tentacle 111 may be constructed to include a compatible material, particularly a high friction elastomeric material, i.e., Tool Dip. TM Coating with this agent restored the desired high coefficient of friction on the gripping surface.

[0036] 1B is a schematic diagram illustrating a tentacle 111 of the soft robotic gripper of FIG. 1A, according to some embodiments of the present disclosure. A palm base plate 102 is positioned at the tentacle base 113 of the tentacle 111, the palm base plate 102 contacts the tentacle base 113 of a lower tube 118, and an upper tube 119 includes a tip 116.

[0037] 1A and 1B, it is contemplated that the outer surface OS of the multiple tentacles 111 and palm base plate 102 of FIG. 1A may be uniquely designed for each user application. For example, the objects may include one or a combination of non-uniform or uniform shapes, different deformabilities, different flexibility, different strengths, sizes, sensor loads, chemical resistance, frictional properties, textures, and the like. These different types of objects present challenges to conventional gripper systems to grasp or move the objects, resulting in a need to find new tentacle structures that overcome these challenges. Some embodiments of the present disclosure are designed such that the outer surface OS of the tentacles 111 and palm base plate 102 overcome these challenges that were not overcome by conventional gripper systems. The innovative tentacle 111 and palm base plate 102 design of each is specifically configured to provide a unique gripping surface to enhance the level of grasping and moving the various types of objects described above compared to the level of grasping the same objects by conventional grippers. Some unique gripping surfaces may include exterior surfaces with one or a combination of dimples and protrusions, surface portions with some roughness and smooth surfaces, molded extensions (rods, horizontal surfaces, etc.), engineered 2D or 3D configurations, various design patterns of dimples or protrusions or both. Additionally, some exterior surface designs may have one or a pattern of ridges, rough surfaces, differently shaped ridges / bumps / indentations, etc., which can be configured to increase one or a combination of gripping force / power, level of grip, etc.

[0038] 1C is a schematic diagram illustrating some components of the soft robotic gripper of FIG. 1A, according to some embodiments of the present disclosure. For example, the spacer 112 is positioned in the center of the ring 114, resulting in a guide ring 116. The lower tube 118 and the upper tube 119 both include the guide ring 116 along a central axis CA. The lower tube 118 and the upper tube 119 are connected by a tube connector 141, and the upper tube 119 has a tip 116 fastened to a distal end of the upper tube 119.

[0039] An aspect of each tentacle is that sensors (not shown) may be implemented along the lower tube 118, the tube connector 141, the upper tube 119, and the tip 116. For example, it is contemplated that the spacer 112, the ring 114, and the tip 116 may have embedded sensors (not shown). Several types of sensors may also include, by way of non-limiting example, real-time sensors, which may include (a) cable tension, (b) cable mobility, (c) wear sensors operable for cables, frayed or broken cable strands, etc., (d) angular rotation measurements, (e) pressure sensors, (f) force sensors, (g) temperature, (h) moisture, (i) friction coefficient sensors, (j) surface conductivity sensors, i.e., to determine if fluids are leaking from the object, such as meat juices, molten fluids from a nominally frozen object, or other fluids from the workspace environment (i.e., water, oil, grease, solvents, deposits from soot, etc.). These sensors can provide data to pass the measured sensitivities back to the soft robotic gripper system (i.e., robot control computer 362, control module 360, sensor control 373, tentacle sensor 366, etc., of FIG. 3E) for aspects related to operation, care, maintenance or any other concerns related to system 300G of FIG. 3G or any component or other system with which the gripper system may be integrated. Additionally, the sensors can be configured to generate sensor data that can be applied to a tentacle state function to determine the state of the tentacle. Tip 116 can also include one of the following sensors: an inertial sensor, a microelectromechanical system device, an accelerometer, a magnetometer, a tactile sensor, and an electromagnetic position tracker.

[0040] Some embodiments may be configured with a distal joint sensor implemented in each tentacle joint or in the distal joint of each tentacle. The term "joint" may include hinge-type joints, ball-and-socket joints, sliding rail joints, rolling wheel-type joints, and elastomeric flexible joints such as the flexible lower tube 118 and upper tube 119. The joint sensor may be configured to generate data that may be applied to a joint state function to determine the state of the distal joint, and the distal joint sensor may be of a type that is a distributed shape sensor or a linear displacement sensor. The linear displacement sensor provides data about the cable associated with the distal joint, and each linear displacement sensor provides data for determining the distance traveled by the cable transmission, and when the distance of all the cables is determined via the linear displacement sensor, the distance of all the cables will indicate the movement of the distal joint. Furthermore, each linear displacement sensor may include one of a linear voltage differential transformer, a Hall effect sensor and magnet, a sliding potentiometer, and a reference element coupled to the cable transmission at a reference point, thereby measuring the movement of the reference element as an indication of the movement of the cable transmission. It is intended that an actuator cable wear detection sensor detects wear on the actuator cable using a measurement of the actuator cable's electrical resistance when the gripper is not gripping an object, and a resistance measurement sensor on the same actuator cable measures grip force using the cable as a strain gauge resistive element when gripping an object. Additionally, the sensor data can be from environmental sensors located in proximity to the tentacle, or sensors such as cameras, video cameras, speed sensors, light sensors, distance sensors, or depth sensors.

[0041] FIG. 1D is a schematic diagram illustrating a cable configuration of the lower tube spacer and ring 116 of FIG. 1C of the soft robotic gripper of FIGS. 1A and 1B according to some embodiments of the present disclosure.

[0042] 1C and 1D, the lower tube ring 114 of Fig. 1C includes the cable path and cables A1, A2, A3, A4 of Fig. 4D that control the flexing of the lower tube 118 of Fig. 1C, and the lower tube spacer 112 of Fig. 1C includes the cable path and cables D1, D2, D3, D4 of Fig. 12D that do not change the flexing of the lower tube 118 of Fig. 1C but provide the flexing of the distal upper tube 119 of Fig. 1C. Path D9 of Fig. 1D and E9 of Fig. 1E can be used for central wiring associated with the sensors in the upper tube 119 of Fig. 1C, the lower tube 118 of Fig. 1C, and the end cap 116.

[0043] FIG. 1E is a schematic diagram illustrating a cable configuration of the upper tube guide ring of the soft robotic gripper of FIG. 1A and FIG. 1B according to some embodiments of the present disclosure. With reference to FIG. 1C and FIG. 1E, the upper tube 119 of FIG. 1C has a guide ring 116 with cable paths F1, F2, F3, F4 for cables that control the bending of the upper tube 119 of FIG. 1C (i.e., the cables actuate the upper tentacle section and terminate at the distal guide ring of the upper tube near the cable paths F1, F2, F3, F4). The upper tube spacer 112 can include cable paths E1, E2, E3, E4 that can provide a path to a second connector attached to the third tube section. Additionally, the central hole E9 can be utilized for wiring of a sensor or other operating device that can provide data related to the end cap 116 of FIG. 1C.

[0044] FIG. 1F is a schematic diagram showing a cable configuration exiting the distal guide ring of the lower tube spacer 112 (C1-C4) in some embodiments of the present disclosure, where the cables are fixed to actuators A-L of FIG. 3A, pass through the openings in the palm base plate of FIG. 3A through a path in the lower tube spacer 112, enter tube connectors 141 (D1-D4), exit tube connectors 141 (E1-E4), enter upper tube guide ring portions 114 (E1-E4), and exit ring sections 114 (F1-F4) of the upper tube guide ring.

[0045] At least one realization was to separate the lower tube cable controlled transmission system from the upper tube cable controlled transmission system. It was observed from the experiment that the lower and upper cable controlled transmission systems had very little operating friction of the tentacle cable transmission mechanism. The lower tube 118 and upper tube 119 sections and the tube connector 141 had a minimum bend radius and no kinking or other constrained movement was observed. However, the realization was preceded by many failed test experiments from which many aspects were learned. For example, one failed test design positioned the lower and upper cables together in one channel or hole, resulting in very high friction and very poor bending performance. Another failed test design positioned the lower cable in one hole in the lower tube and the upper cable in another hole in the lower tube, and the upper cable was inserted from the central through hole of the connector into an upper guide ring that included individual through holes for each upper cable for the upper tube. However, these failed test designs also exhibited very high friction and very poor bending performance. Thus, what we learned from these failed test designs is that at least one challenge to overcome in developing embodiments of the present disclosure must be figuring out how to isolate the lower cable transmission system from the upper cable transmission system.

[0046] Continuing to refer to FIG. 1F, it eventually became apparent that even though there was enough physical space in the hole for the cables to extend together, as soon as the cables were tensioned and the tentacles were bent, the cable tension pulled all of the cables to one side of the hole, causing the cables to rub against each other. Whether the cables were nylon coated or uncoated, or smooth steel cables were used, it was observed that the extremely high coefficient of friction that caused the cables to essentially stick together placed unacceptable friction-induced loads on the servo motors, resulting in poor performance and other adverse effects on the test system. It was further observed that some cables would sag and then loop under a second cable, which when subsequently tensioned would fix the position of the first cable, causing adverse effects on the tentacles and in some cases failure of the gripper operation.

[0047] Another realization gained later was that by designing routing holes for each cable in both the spacer 112 and ring 114 of the guide ring 116, tentacle performance could be improved with high speed operation, minimal friction, and minimal load on the servo motor. Another innovative feature incorporated in the innovative guide ring for some embodiments includes using low friction coefficient materials for the spacers and rings, i.e., materials with very low friction coefficients when placed in close proximity to uncoated or coated cables. Improved performance and increased tentacle grip strength in very cramped tentacle bending positions / applications have been observed from test experiments. It is believed that at least one reason for the increased tolerable grip strength is due to the reduced friction coefficient caused by the spacer and ring cable routing compared to conventional cable designs and configurations for the same size conventional gripper system. It is intended that the spacer and ring material types may be selected for each application requiring one or more of higher friction coefficient type materials, higher friction resistance type materials, higher strength type materials, and low weight type materials. It is further contemplated that small or minimum size bushings or bearings can be used as inserts between the spacer or ring and the cable, and the bushings / bearings can be designed to have a specific coefficient of friction material that may be determined by the required operational goals intended by the operator.

[0048] With continued reference to FIG. 1F, further experimental findings showed that test cables encased in a sheath, such as a bicycle brake cable, exhibited insufficient flexibility, performing poorly and failing to meet the flexibility performance goals of the present disclosure.

[0049] 2A, 2B, and 2C are schematic diagrams illustrating a test experiment of a transmission channel configuration that resulted in an optimized transmission channel for an elastic connector 241 that minimizes cable friction and cable material fatigue, according to an embodiment of the present disclosure. FIG. 2A shows a side view of the test cable paths 204 and 206 of a failed test path and a side view of the innovative cable path used in an embodiment 208 of the present disclosure for an elastic connector 241, FIG. 2B shows a perspective view of FIG. 2A of the failed cable path and the innovative path, and FIG. 2C shows a top view of FIG. 2A.

[0050] With regard to fabrication of the elastic connector 241, test cases A1 and A2 can be fabricated with a simple drill bit, however such channel configuration resulted in sharp corners 204, 214 in Figures 2A and 2B that increased cable friction and resulted in damage to the cable material during cable operation. Due to the high cable friction and damage to the cable material, test cases A1 and A2 were not desirable for the tentacle gripper design and were not tested further.

[0051] 2A-2C, in test cases B1 and B2, the inlet and outlet openings of the transmission channel were incorporated with radius curvature 206, 216 of FIG. 2A and FIG. 2B. Testing revealed that the addition of this radius curvature reduced cable friction and cable material fatigue during cable operation. However, the incorporation of the radius curvature required advanced manufacturing techniques such as three-dimensional (3D) printing to generate the 3D curve of the preferred transmission channel design shape, or injection molding using notches. Further testing later revealed that although the addition of the radius curvature reduced cable friction and cable material damage, the observed cable friction and cable material damage, as well as cable material fatigue, were at a level that could not provide sustainable cable friction and cable material performance to meet the desired goals, performance aspects, and requirements expected of the embodiments of the present disclosure. Test cases B1 and B2 revealed that the radius curvature had to be optimized to substantially reduce cable bending and cable wear during tentacle motion, thereby minimizing cable material fatigue during tentacle gripper operation.

[0052] Optimized transmission channel inlet / outlet opening cable routing 2A-2C, an embodiment of the present disclosure includes optimized transmission channels based on a mathematically constructed formula that further increases the radii 208, 218 of FIG. 2A and FIG. 2B of the structural cable path design for the inlet / outlet opening of each transmission channel. The formula includes obtaining a measured distance between the guide disk through hole, i.e., the through hole through which the cable enters and exits, and the inlet / outlet transmission channel opening of the elastic connector 241. The reason for measuring this distance is to use this measured distance together with a pair of circular arcs of constant and equal radius to construct an optimal cable path for the inlet / outlet transmission channel opening. Maximizing the radius of the pair of circular arcs within the measured distance minimizes the cable side load resulting from the cable movement, and at the same time minimizes the transmission friction and cable bending fatigue during the operation of the tentacle gripper.

[0053] The formula for optimal cable routing The formula for constructing the optimum cable path depends on the axial distance (AD) in FIG. 2A between the inlet / outlet openings measured longitudinally along the tentacle, and the radial distance (RD) in FIG. 2A between the inlet / outlet openings measured parallel to the long axis of the tentacle. The optimum cable path starts at the inlet opening location, proceeds along a circular arc from the original inlet opening toward the distal end of the tentacle to a location AD / 2, and reaches the location at an angle equal to the arctangent of ((AD / 2)÷(RD / 2)). The optimum cable path continues along a circular arc from this location to the outlet opening, reaching the outlet opening at an angle of zero. The cable path diameter can be slightly larger than the cable diameter plus a minimum diameter that allows for lubrication.

[0054] 2A-2C 208 and 218, this formula is mathematically constructed to create an optimized radius curvature at the inlet / outlet transmission channel openings that maximizes the radius of the two arcs in the transmission channel design, resulting in minimal cable bending and cable wear in the cable path while the cable is moving or during tentacle movement, thereby minimizing cable material fatigue during tentacle gripper operation and at the same time minimizing cable side loads caused by cable movement through the inlet / outlet transmission channel openings of the transmission channel, and at the same time minimizing transmission friction during tentacle gripper operation, thereby increasing the operating life of the cable and therefore the tentacle gripper.

[0055] Through-hole-transmission aperture (PH-TA) cable routing In developing the optimized transmission channel entry / exit opening cable path, it was found from the experiments that there were further problems / challenges that had to be overcome in terms of designing a cable path with minimum cable friction and minimum bending path between the guide disk through-holes, i.e. ring through-holes or spacer through-holes, and the entry / exit transmission channel openings. In addition, it was recognized that there would be less cable friction if the lower and upper cables did not cross in the central through-hole of the elastic connector 241. This could be achieved by ensuring that the four openings proximal to the tentacle base, i.e. palm base plate, that enter the central through-hole of the elastic connector 241 are connected to the exit openings at the distal end of the elastic connector 241, and that the four openings distal to the tentacle base, i.e. palm base plate, that enter the central through-hole of the elastic connector are connected to the exit openings at the proximal end of the elastic connector. Thus, in the optimized design, the entry openings for the cables passing through the spacer through-holes are positioned such that they should enter the central through-hole closer to the tentacle base than the exit openings.

[0056] 2A-2C 208 and 218, the realization gained from substantial testing experiments is that the elastic connector 241 needed to include lower and upper portions of the same length so that the lower and upper portions of the elastic connector could extend into the lower and upper elastic members and contact each guide disk, i.e., the distal guide disk of the lower elastic member and the proximal guide disk of the upper elastic member. This would minimize the cable path distance between each guide disk and the lower and upper portions of the elastic connector. Therefore, this cable path is referred to as a passthrough hole-transfer aperture (PH-TA) cable path that is configured to align the centers of the guide disk through holes, i.e., ring through holes or spacer through holes, with the lower and upper portions of the elastic connector. This cable path has the minimum cable path distance and exhibits low cable friction and minimal bending path when tested in combination with the transfer channel cable path aperture embodiment.

[0057] FIG. 2D is a schematic diagram showing a top view of a guide disk having a passage with rounded or convex edges and a central through hole to minimize friction and wear on the elastic tube, according to some embodiments of the present disclosure.

[0058] 3A is a schematic diagram illustrating a soft robotic tentacle gripper system 340A according to some embodiments of the present disclosure. For example, gripper system 300A includes three tentacles 311A, 311B, 311C actuated with twelve servo motors A-L having a range of motion of 180°. Each of the twelve servo motors A-L has its control lines connected to separate digital I / O pins on the Arduino Mega so that the position of each servo motor can be independently set to any value between 0° and 180°. Each servo motor A-L holds two flexible nylon-coated stainless steel cables or cables 329A, 329B, 0.92 mm in diameter, attached to either end of a bell crank. Each servo motor cable pair bends the same tentacle section in opposite directions, and a 90° angle on a servo motor A-L is nominally "zero curvature" for that degree of freedom on that tentacle 311A, 311B, 311C, such that rotating the servo motor shaft toward 0° bends that tentacle section in one direction, and rotating the shaft toward 180° bends that same tentacle section in the opposite direction. Other components of the gripper system 300A may include wiring 331 connecting the servo motors (actuators) A-L, a controllable palm base plate 302A, and a controller 350. The controller 350 may be connected to a centralized control system (see 300G in FIG. 3G), or the centralized control system may be incorporated in the same location as the controller 350, or located in some other location, or may be integral to the centralized controller 350, depending on the user requirements in terms of configuration and application.

[0059] Gripper system 300A is fully actuated via servo motors A-L, moving independently in all degrees of freedom, achieving over + / -120° of bending per tentacle section versus + / -90° of motion on the servo motor shafts (this motion is limited to 120° versus the tentacle sections being limited to 162° due to reduced available cable motion due to limited bellcrank arm length, not due to lack of servo torque).

[0060] Eight predefined grasps are programmed into the Arduino's Electrically Erasable and Programmable Read Only Memory (EEPROM) as an example, with many more grasps to be programmed. These grasps may include several types of pinch and wraparound grips (see Figures 5A-6B). The operator can select among these stored grasps using potentiometers, and a serial-over-USB command line interface allows for full independent control of the individual servo motors by human or control software. The positions of each servo motor, actuator A-L, are stored in an array, and if the operator / user wants to generate another preprogrammed grasp, they can do so by simply saving the current array with a unique name. Using the Arduino's EEPROM, all stored arrays can be recalled, edited, and resaved at any time. The total current draw and voltage delivered to the gripper system 300A is less than 8 amps (at a nominally constant 6.2 volts) and is monitored in the power supply (see 1008, 1029 in Figure 10).

[0061] Cable-controlled transmission system Experiments have found some realizations of the cable-controlled transmission system for some embodiments of the present disclosure. All experiments included active control using sets of opposing cables 329A, 329B (e.g., one set to control "pitch" and an orthogonal set to control "yaw").

[0062] Motors / Actuators: A soft robotic actuator or motor can be connected to the cable transmission system of the tentacle, and the tentacle component material may affect the performance of the motor / actuator. For example, the tentacle component may be formed of an elastomeric material such as rubber, or a thin wall of plastic arranged to cover the internal components such as the spacer and tube connector structure, and when a force is applied to the cable by the actuator, the tentacle component may stretch and / or bend under the pressure applied by the actuator. At least one operational benefit or advantage of this motor / actuator compared to other types of force generating devices is that the position of the motor that moves the cable of the cable transmission system can be soft-fixed by servo control to hold the position of the cable and prevent the movement of the lower or upper tube. This fixing aspect provides even higher tentacle operation performance, for example by holding the position of the lower tube spacer.

[0063] Centralized Controller: Several centralized controller experiments were found to be tested with embodiments of the present disclosure and later incorporated into several embodiments. Some of them included a centralized controller programmed with stored logic, the logic including instructions to complete the steps of (a) detecting motions having human-like characteristics from an operator wearing a teaching glove having sensors to obtain respective sets of command instructions corresponding to stored construction objects, (b) generating one or more signals by the sensors representing at least a portion of the motion, (d) converting the one or more signals into a set of command instructions that actuate the tentacles or actuate the tentacles and other devices in a network of robotic devices (the set of command instructions is functionally equivalent to an action), and (e) communicating the set of instructions to a centralized control system and a network of robotic devices (the robotic devices being robotic assembly devices configurable to operate with a robotic system).

[0064] Continuing to refer to FIG. 3A, other aspects incorporated into the data used to determine the command database include various grasping actions by the multiple tentacles. For example, some grasping actions may include (a) an ulnar or palmar grasp (i.e., a grasp that is a raking action, with the fingering apparatus holding the object against a palm-like surface and trapping it) and a raking grasp (i.e., the fingering apparatus grasps the object, but the fingering apparatus does not have a thumb and the fingering apparatus does all of the holding). Based on this experimentation, several grip sequences of motion commands were developed to move different shaped objects.

[0065] Another realization that has been made is that a centralized controller for a soft robotic gripper must overcome the shortcomings of conventional gripper control architectures by recognizing the soft gripper control variability that can vary and be very different depending on the differences in the various underlying gripper material properties. For example, conventional gripper control architectures have grip force problems because they are unable to generate large gripping forces due to the types of materials used in the construction of the soft robotic gripper structure. Some of the reasons for grip force problems with conventional gripper control architectures may be due to the lack of quantitative analysis in identifying the different grip strengths associated with each grip by a gripper gripping an object. Other reasons for grip force problems may be due to the failure of quantitative analysis in identifying the individual advantages of each grip mode from a large number of different grip modes, and the failure of quantitative analysis in identifying the individual advantages of each pull direction from a large number of different pull directions.

[0066] Continuing to refer to FIG. 3A, for example, the command database includes multiple sets of coordinated control command instructions or tentacle motion programs that were developed based on the quantitatively evaluated tentacle performance and fully independent actuation of the tentacles used to develop any gripping scheme, gripping mode, gripping strength, and load capacity used to build a coordinated set of control command instructions or multi-tentacle motion program that can be later accessed by the centralized controller or its components. Any gripping scheme can be an internal grasp and an external grasp, each of which can be based on friction (i.e., pinch or clamp) or based on kinematic constraints (i.e., an enveloping grasp or a power grasp, like a human hand grasping the handle of a baseball bat, that can maintain a grip even in the absence of friction). An external grasp has a tentacle positioned on the outside of an object and pushed inward, and an internal grasp has one of the tentacles inserted inside the object, i.e., into an orifice, hole, etc., and then the tentacle bends or expands (distends) to grasp the object. Further, the gripping modes may include modes in which one tentacle braces against or on another tentacle, increasing the force on the braced tentacle and thus increasing the grip strength. During the predefined test period, the test central control system may set a range of minimum gripping forces while ensuring a maximum gripping quality that may be predefined based on predefined gripping thresholds. For example, some of the predefined gripping thresholds may include measured quantities obtained from real-time sensor data, such as, but not limited to, the amount of pressure on the tentacle, i.e., on points along the tentacle, including points such as joints, the measured angle and velocity of the tentacle, the amount of slippage obtained from the change in the amount of pressure over time, and the amount of force on the tentacle.

[0067] As described above, the centralized controller can access an object database containing an indexing of each object identification and object pose, and a command database for the desired type of motion to retrieve time sequences of commanded actuator positions and torques, as well as ranges of expected actuator forces and expected joint torques. The desired type of motion can be bending along any axis of each tentacle, including X, Y, Z, length and axial rotation. The ranges of expected actuator forces and expected joint torques are well within the range achievable by elastomeric tentacles with cable tension elements.

[0068] 3A , specifically, some other aspects of the quantitative analysis include grip types for gripping different shaped objects and grip types for gripping different object poses of multiple poses for each shaped object. The grip types for both object shape and pose may include aspects such as (1) grip strength for each tentacle to hold the object shape and overall grip strength for all tentacles to hold the object shape, (2) grip performance for each grip mode of multiple grip modes, and (3) pulling performance for each pull direction of multiple pull directions.

[0069] Some other aspects incorporated into the data used to determine the command database include gripping force control variability based on the type of gripper material properties, the gripping performance of multiple tentacles to move a particular object configuration, the sequence of control signals executed including actuator positions, the expected torque, force and position expected to be experienced at those steps, the timing between steps of the sequence of steps, and an alternative set of steps to take at each step if the initial torque, force or position identified, or any other measurement, was not within a pre-defined range. The gripping quality can be a measure of the relationship between the distributed force and moment on the gripped object that can be applied by the gripper to avoid damaging the object. One pre-defined test included moving the object to determine the best quality grip based on a pre-defined gripping threshold. The higher the level of disturbance possible, the higher the quality of the grip. For example, the total gripping force of the tentacles can be considered as the sum of the individual forces. Other forces that may be considered include static forces with respect to the direction of gravity (i.e., may include maximum acceleration in the direction of gravity) and tentacle acceleration perpendicular to gravity (i.e., may instead be detected based on measurements of slippage).

[0070] Types of sensors and applications using sensor data 3A, it is contemplated that environmental sensors positioned within the working area of ​​the object to be moved can obtain environmental data (i.e., moisture, temperature, dust, etc.) to assist the tentacles in grasping and releasing the micro-objects. For example, the environmental data can include using measured data such as measured capillary forces, electrostatic forces, etc. Moisture measured from the air within the working area condenses into a thin layer of water on the surface of the object that the tentacles will use to move the object.

[0071] Development of the embodiment Continuing to refer to FIG. 3A, in developing the embodiments of the present disclosure, some experiments included both inspiration from natural evolution and human anatomical investigations to come up with some mechanical structures, including tentacles, and control capabilities via a controller with a processor and other components. In some experimental aspects, sea anemones, octopuses, squids, and elephants were analyzed and investigated. All of these included flexible manipulative organ postures, but the control methods were fundamentally different. Some of the human-designed tentacle experiments included multiple actuation methods (pneumatic, hydraulic, tension cable) based on a single tentacle with a few actuators. However, at least another realization gained from these experiments is that a single tentacle with a few actuator configuration is too limiting in many aspects. Therefore, another realization is that the soft gripper of the present disclosure needed to increase the utility of the tentacles over a single tentacle configuration, such that each tentacle has the ability to be independently actuated, along with a centralized control that coordinates the movement of all tentacle gripping actions.

[0072] Thus, the command database can be used to identify an associated set of coordinated control command instructions or object movement programs based on identifying the object shape, the pose of the particular object shape, and receiving a particular type of multi-tentacle movement action via the operator. For example, to use the command database of stored sets of coordinated control command instructions or object movement programs, the object shape needs to be identified and a pose specific to the object needs to be identified. However, if the object shape is circular, there is only one circular pose, so there is no need to incorporate the object pose. For object shapes with multiple object poses, in order to identify a set of coordinated control command instructions from the command database, it is then also necessary to identify the specific object pose associated with the object shape.

[0073] 3B is a schematic diagram illustrating a configuration of a fixable or controllable palm baseplate 302B according to some embodiments of the present disclosure, where each tentacle of a plurality of tentacles can move on the palm baseplate 302B in the X direction along the X axis and in the Y direction along the Y axis independently of the other remaining tentacles. For example, each tentacle can be movable using a screw mechanism that translates the tentacle's lower elastic tube or proximal end along the X direction or the Y direction or both of the palm baseplate 302B to further optimize the strength performance of the tentacle group. Commands are generated by a centralized control system and executed by an actuator (not shown) that actuates the screw mechanism.

[0074] 3C is a schematic diagram illustrating another configuration of a controllable palm baseplate 302C according to some embodiments of the present disclosure, the controllable palm baseplate includes controllable outer edges 1C, 2C, 3C, 4C that can be adjusted in or out along the horizontal plane B (HP-B) or along the extent of the vertical plane A (VP-A) of the controllable palm baseplate. For example, tentacles 1, 2, 3, 4, 5 can be controlled utilizing any one of the controllable portions 1C, 2C, 3C, 4C of the palm baseplate 302C to further optimize the strength performance of the tentacles.

[0075] FIG. 3D is a schematic diagram showing a controllable palm base plate according to some embodiments of the present disclosure, which is adjustable along the Y axis such that it can be tilted towards or away from tentacles 301D, 302D, 303D, and can be adjusted left and right along the X axis to optimize the strength performance of the tentacles.

[0076] FIG. 3E is a schematic diagram showing a side view of a detachable partial grab device according to some embodiments of the present disclosure. The partial grab 361E provides additional surface area for the tentacles to grip and conforms closely to the contours of the object, resulting in a higher level of secure gripping, an increased amount of secure and stable gripping, and a maximum gripping quality with a lower rate of damaging or dropping fragile objects. It was observed from the test experiments that the partial grab performed substantially better when it had a reinforcing support rib portion attached to each tentacle and palm base plate and axially aligned with several lower member guide rings. It was also observed from the test experiments that both the amount of gripping performance and the amount of elasticity of the partial grab should be customized according to a set of predefined optimized gripping thresholds corresponding to a particular object configuration and associated features / characteristics. For example, part of the customization of the partial grab included one or more gripping features located between each tentacle that contacts the object to be moved. Grip features can be, for example, the amount of indentations, protrusions, etc., the type of pattern of indentations, protrusions, etc., the amount of one or more textured materials, the pattern of one or more textured materials, the amount of sticky or non-stick material or pattern, the size of spaces in a mesh material, or the amount of abrasive material.

[0077] Test experiments were designed to move test objects in different work environments in different work areas using the tentacles of the present disclosure equipped with partial grabs, each of which was customized for each test object and the conventional gripper of the conventional gripper system. The tentacles and the tentacles equipped with partial grabs had capabilities including (a) moving objects with OD of about 30-95 mm with a gap between opposing fingers of about 63.5 mm, (b) a cycle time of 88 msec to close and 166 msec to open, and (c) a gripping force of about 2.5 lb per finger. Each test object used in a test experiment had known data stored, including one or more sets of known, pre-determined, optimized gripping thresholds corresponding to the configuration and associated features / characteristics of the test object. Additionally, each partial grab equipped on the tentacle was customized for each test object for each test experiment.

[0078] Partial grab 361E was tested to grip and move objects and compared to the tentacles of the present disclosure and conventional gripper systems. The objects included objects having sizes ranging from 425mm long x 200mm wide x 200mm high, objects classified under the GS1-GPC list of international commodity classifications, objects that are deformable, non-deformable, and objects with different configurations, features, and characteristics as disclosed. The tests included obtaining real-time sensor data from real-time sensing of environmental sensors and sensors associated with the gripper system of the present disclosure. Additionally, the test embodiment used food contact materials that meet the requirements of FDA-CFR Chapter 21, was rated IP67 ingress protection, and operated at decibels below 85dB. During the tests, it was expected that some objects with uneven shapes, deformable objects, etc. would fall off the tentacles because the tentacles had not acquired initial quality for gripping.

[0079] It was observed from the test experiments that the performance of the tentacle of the present disclosure was superior in all areas compared to a conventional gripper system of the same size. It was also observed that the performance of the tentacle fitted with the partial grab 361E was superior, providing the observer with an increased level of confidence in the tentacle's high quality grip in a given test experiment, and when the tentacle fitted with the partial grab was compared to a conventional gripper system of the same size, the former outperformed in all areas. The advantage of using at least one partial grab is that customized configurations of the partial grab 361E are produced at minimal cost, given the increased level of reliability and grip quality performance discussed above. Other additional advantages include faster tentacle operation times, which translates into less manufacturing time per unit, reduced maintenance costs, and reduced material fatigue of the tentacle parts. It is contemplated that partial grab 361E may be configured to have detachable portions, such as to be attached to a palm base plate while contacting only two tentacles of a three-tentacle gripper, all depending on the particular object being intended to be moved given the object's type, characteristics and features.

[0080] The partial grab 361E may be constructed of one or more materials having one or more features such as a level of stiffness, a level of flexibility, a level of gripping ability, a level of durability, a level of texture, a level of stick resistance, varying the amount of open space in the partial grab structure, a pattern of open space, a level of transparency, a level of temperature rating, a level of strength, a level of chemical resistance, a level of partial grab material fatigue, an amount of durability, an amount of thickness, etc. The amount and weight of the cost of the material of the partial grab may depend on the user's operational requirements or some other parameters specific to the user. The partial grab 361E may include a pattern of open space based on a predefined optimized test performance for moving a test object. It is contemplated that an operator may provide operator data in real time before the tentacle moves an object, the data provided by the operator indicating that a customized partial grab should be attached to the tentacle to move a particular object to obtain optimal performance features. The operator data can be compared to a stored database having recommended operator customized partial grabs to attach to the tentacles, and the centralized control system can generate a set of commands to obtain the customized partial grabs to attach to the tentacles before moving the object.

[0081] FIG. 3F is a schematic diagram showing a top view of a detachable partial glove according to some embodiments of the present disclosure. The detachable or fixed partial glove 361F can be attached to the palm base plate 307F in a snap fashion, and the partial glove 361F can be attached to the tentacles 301F, 302F, 303F in a conventional manner, Velcro®, straps, or other manners. The partial glove 361F is attached to the palm base plate 361F and extends along the length of each tentacle. As previously mentioned, the length of each partial glove may be pre-determined. The partial glove 361F wraps around the contours of each object of different shapes and sizes to provide a large surface area to hold and cover the object being moved. For example, pick-and-place type applications may benefit most from customized grab-mounted tentacles, including applications involving moving types of objects such as organic foods, fragile objects, and the like. Partial grab 361F provides increased grip control and grip quality and reduces production time per unit object movement compared to conventional gripper systems of the same size.

[0082] Technical properties and practical applications The disclosed embodiments are configured to be compatible with most commercially available robot controllers in gripper and non-gripper industries, providing turnkey integration to allow users / operators to quickly install and operate the innovative multi-tentaclerial system for new or retrofit applications without additional components or control hardware. The disclosed embodiments are configured to operate in a wide variety of gripper and non-gripper industries. For example, the innovative embodiments can perform tasks at a faster pace than traditional gripper systems due to the new tentacle design and controller nature. Gripper industries where the innovative embodiments outperform traditional gripper systems include collaborative human / gripper systems where the innovative embodiments collaborate with humans in agricultural harvesting, automated fruit and vegetable harvesting and packaging. The innovative embodiments outperform traditional gripper systems. The reasons are (a) high dexterity of multiple tentacles, (b) low cost, physical robustness and human-safe operation, completing tasks with high productivity and human safety, (c) little or no product loss in product handling, performing highly sanitized and delicate operations, and (d) multi-tentacle gripping with 12DoF and optional null space provides high performance for sorting and packaging applications involving delicate produce and baked goods such as apples or small chocolates, as well as other similar delicate storage products. Other gripper industries include the medical, warehouse and manufacturing industries, where the innovative multi-tentacle system (a) can provide high medical support, and (b) can perform well in e-commerce warehouses that require high dexterity grippers capable of manipulating packages of various sizes, shapes, weights and softness (i.e. bags filled with products), as well as warehouse logistics applications involving objects of unpredictable sizes.

[0083] Some advantages and benefits of the disclosed embodiments of operating the above gripper system operating environments over conventional gripper designs and systems include: (a) fewer component parts to operate, thus eliminating bearings, shafts, and other mechanisms that may fail and / or require extensive maintenance costs in some gripper industries in harsh environments such as dust, smoke, etc.; (b) lighter weight, improved safety due to lighter weight, and high dexterity, minimizing impact forces upon impact; (c) adaptable, plug and operate, repeatable and reliable with quick release tentacle mechanism option; (d) configured to operate and complete advanced assembly tasks that currently require large labor forces; (e) packaging, food and beverage, and industrial robots adaptively handle various objects with a single easy-to-use device, eliminating the need for tool / tentacle changes and complex vision / sensor requirements; and (f) millisecond response fast controller ensures repeatable and reliable operation.

[0084] FIG. 3G is a schematic diagram illustrating some components of a centralized controller system 300G of a soft robotic gripper system according to some embodiments of the present disclosure. For example, the centralized control computer 340 may be connected to sensors 330 positioned within a sensing distance of at least one object (not shown) moved by the centralized control computer 340 located in the environment (not shown). The other sensors may be sensors embedded in one or more tentacles, or sensors attached to the robotic system, or sensors located in areas of the workspace. Some of the other sensors may include cameras, video cameras, speed sensors, light sensors, environmental related sensors (i.e., temperature, humidity, fire, air, water, barometer sensors, etc.). The other sensors may obtain object data, including data from imaging or optical sensors that can determine physical object features such as size, shape, position, orientation, exterior texture, porosity, and object gripping capabilities based on observing the exterior of the object. Some of the other sensors may include distance sensors, depth sensors, force sensing, tactile sensing, pressure sensing, voltage sensing, conductivity sensing, ultrasonic sensing, x-ray sensing, or other sensing, such as to determine physical aspects of the grasped object. Other sensing data may be obtained in association with the deformability and weight of the grasped object. Sensor 330 and the other sensors may be hardwired and / or wireless and may be connected to a network 332 that is within an operating distance range of sensor 330 and the other sensors.

[0085] The centralized control computer 340 is connected to a bus system 322, which in turn connects to a storage 342. The storage 342 may include an object database 344 and other databases 346. A memory 348 is connected to the bus system 322 along with a receiver interface 351, a receiver 352 and a power supply 355. The power supply 355 may be connected to the bus system 322, connected to the control interface 353, or configured to have numerous other configurations, all as intended by a particular user / operator for the specific operational needs. The power supply may be electrical, a renewable energy source, chemical, may be directly connected to the soft robotic gripper system, may have multiple external power sources of different types, or both. A transmitter interface 357 and a transmitter 359 are connected to the bus system 322. Also connected to the bus system 322 is a network interface controller 350 (NIC) (also known as a network interface card, network adapter, LAN adapter, or physical network interface, and similar terms, which is a computer hardware component that connects a computer to a computer network) which may communicate with the network 332 or other different types of wireless networks (not shown), again depending on the particular needs of the user / operator for the performance and sensitivity required for their intended use.

[0086] 3G, the control module 360 ​​may be connected to the bus system 322. The control module 360 ​​may be connected to a robot control computer 362, which helps to direct the components of the robot control computer 362. Some components may include, as non-limiting examples, multiple tentacles 364, tentacle sensors 366, a robot base system 368, and other robot related components 369, all of which may be connected to the robot control computer 362 and / or the control module 360 ​​via 371. The control module 360 ​​may be configured with the robot control computer 362 to confirm or deny grip of an object (not shown) by checking the force state of the multi-tentacles once and reporting back, or to continuously monitor the multi-tentacle motion state at a frame rate and interrupt the robot control if the multi-tentacles indicate a good grip or an ineffective grip. It is contemplated that the robot control computer may have a power module 374, which may include any power source that provides power to the robot system, such as a power adapter or a rechargeable battery pack, depending on the user's specific needs.

[0087] The centralized control computer 340 may implement or execute stored instructions stored in memory 348, storage 342, or accessed from cloud storage via network 332; in particular, the centralized control computer 340 may access pre-stored data, including data from object database 344, instruction database 346, or other stored data from other databases (not shown) as described above. It is contemplated that some embodiments of the present disclosure may have a centralized controller designed with a control architecture to obtain and receive data that may be stored in a database and later accessed by the centralized controller. The data for the database may relate to received data from third parties, past operating soft robotic gripper data, sensed data sensed from sensors in the object's environment, and other data related to the operation, maintenance, management, performance of the soft robotic gripper, data related to methods of adaptability to new concepts, etc., or data not related to the soft robotic gripper.

[0088] 3G, the robot control computer 362 can command each tentacle of the plurality of tentacles 364 to move along an X-axis, a Y-axis, or a Z-axis 425 (see FIG. 4) to open or close the plurality of tentacles, or other actions. It is contemplated that the centralized controller system 300G can include a sensor control computer 373 connected to the tentacles 364, the tentacle sensors 366, the robot base system 368, and other components 369. The sensor control computer 373 can include a hardware processor (not shown) connected to a memory (not shown) containing stored software (not shown) and a pre-trained XY set database including labels and surface force values. The hardware processor (not shown) can implement or execute stored instructions stored in the memory (not shown) and, when any sensor data is received, can compare the received data to the stored data and transmit sensor data / signals, such as warning messages, the degree of grasping / grip by one or all tentacles, to the robot control computer 362 or other components of the system 1400.

[0089] It is contemplated that the sensor control 373 of FIG. 3G may be used to determine the location and orientation of the object 496 of FIG. 4 in the bin 496 of FIG. 4. The sensors 409A-D of FIG. 4 may be configured as cameras that generate images for subsequent tentacle grip analysis, such as determining whether a tentacle has a particular type of tentacle grip, whether or not it grasps one or more objects 495 from the bin 496 of FIG. 4, or both, along with the location and orientation of the grasped object. It is contemplated that other malfunctions and errors may occur during the handling of objects by the multi-tentacle gripping system in connection with grasping a bin or any other related operation. At least one reason for utilizing the sensor control 373 of FIG. 3E is that objects may become stuck or entangled, and such entanglement of objects may cause errors if not recognized and removed early. For example, if the tentacle grasps multiple objects, the multiple objects may fall out of the grip or the multiple objects may collide with other components, either scenario resulting in loss per unit cost, lost production time, safety issues for human operators, etc. The sensor locations may be configured based on the intended goal or monitoring type of analysis specific to the operator's intended operational management goals, concerns, and machine system requirements. Additionally, software may be associated with the sensor control system that can use images from the camera to detect the location and orientation of an object in the tentacle after the tentacle grasps the object from the bin 496 of FIG. 4. This software may be configured in a number of different ways, such as being the same as the software that detects the location of the object in the bin 496 of FIG. 4, or being separate from that software. This software may be in a controller that has or is a computing device, or in some cases may be in a separate computing device. It is contemplated that the camera may be a smart camera with a built-in computing device.

[0090] 3E, the control or controller module 360 ​​may include different applications and programs for implementing the methods of the present disclosure. For example, the control module 360 ​​may include applications for processing information received from the robot control computer 362, such as applications for generating graphs, motion-related models including 3D models of the environment in which the robotic system is placed, and object-specific type models. The control module 360 ​​may include applications for interpreting control inputs or input signals received via the input circuitry / interface module 380, storage module 382, ​​and creating commands / actions in the robot control computer 362. For example, some aspects of the robotic system may include various control methods related to stable and scaled motion, control related to combinations of tentacles with palm-based structures or other structures, collision detection and avoidance, compliance control and constrained motion, etc. The input circuitry may integrate display features as depicted in FIG. 9 and FIG. 10. The user display may be configured for a view and model of the object. The display may also be configured to operate to receive control inputs from various input devices (i.e., voice and / or audio recognition software, touch screens, and joystick controllers). Examples of other input devices relevant to human collaboration or collaboration may include, for example, neural and brain implant control, and motion and / or gesture tracking systems. At least some advantages are gained by having an efficient centralized controller architecture design for industrial manufacturing applications, all of which require extensive engineering to integrate the requisite operational theory and practice knowledge with industrial / manufacturing issues that impact stability, safety, and serviceability.

[0091] With regard to the stored databases 344, 346, the data stored in the databases 344, 346 can include tentacle gripper / object manipulations that have been pre-tested during a test period and stored in the database to identify tentacle gripper / object manipulations optimized according to pre-defined tentacle gripper / object motion thresholds to move test objects from a start position to a final position under different environmental conditions in different types of test object workspaces.

[0092] 3E, the object configuration (shape and pose) is first used to identify a stored object configuration from a stored database of object configurations using target data obtained from the gripper system's real-time sensors. However, each object configuration has other aspects including different size, weight, position, orientation, etc., in combination with object properties (i.e., (a) the type of texture of the object's exterior surface material, (b) the object's two-dimensional (2D) or 3D shape, or (c) the object's gripping capabilities, such as wettability, smoothness, or slipperiness).

[0093] During the testing period, each object configuration (i.e., different sizes, weights, locations, orientations, etc.) combined with the above object features are tested with all gripping methods, gripping modes, gripping strengths, payload capacities to grip objects of different sizes, weights, locations, orientations, and object properties (i.e., surface smoothness / texture, stiffness / deformability), and also tested to move test objects from start position to final position for each type of object workspace, i.e., random test objects in a bin, test objects placed on a work surface, etc., to identify the optimized tentacle gripper / object manipulation according to the stored pre-defined tentacle gripper / object motion thresholds under different types of environmental conditions.

[0094] 3E, the test data is recorded in a plurality of databases, the test data being gripping classifiers, object properties and features, tentacle configurations, environmental conditions, object workspaces, and sets of commands (i.e., different object configurations combined with one or more object properties and features) corresponding to all combinations stored in the plurality of databases. Each classification in each database corresponds to data from the other databases. For example, the object database may store first test object configurations having a particular shape, pose, size, weight, position, orientation, etc., along with object properties (i.e., type of texture of the outer surface material of the first test object, etc.). For each type of object workspace, the first test object may include corresponding data such as gripping method, gripping mode, gripping strength, payload capacity, etc., for gripping the first test object under one or more environmental conditions and moving the first test object from a start position to a final position, all of which correspond to the stored sets of commands (i.e., tentacle gripper / object manipulation optimized according to predetermined tentacle gripper / object motion thresholds). Within the first test object stored set of commands in the command database, there is a respective set or subset of commands within the set of commands that uniquely corresponds to each different combination of the object configuration and the one or more first test object configurations.

[0095] Thus, upon receiving operator data and object data from the real-time sensor, the received operator data is compared to the stored data to identify whether the received data corresponds to a first test object, such as a possible pick-up action in the received operator data. If a pick-up action is received and identified in the stored database, a corresponding set or subset of commands for the pick-up action stored in the command database is selected. Furthermore, upon receiving object data from the real-time sensor, the received object data is compared to the stored data to identify whether the received object data corresponds to a first test object, such as a texture type of the outer surface material of the first test object. If the texture type of the outer surface material of the first test object is in the stored database, a corresponding set or subset of commands for the texture type of the outer surface material of the first test object stored in the command database is selected. Note that the tentacles may have one or more tentacle configurations with a separate set of data associated with each tentacle configuration for different combinations of object configurations and object properties and features.

[0096] 3E, it is contemplated that the databases may include an instruction database, an object history database, an instruction history database, an environment database, a human worker database (each worker including a profile associated with the worker's profile and performance characteristics (i.e., worker speed associated with the worker, worker disability limitations, worker health, etc.)), and other databases having data regarding motions and information available to the soft robotic gripper system. Again, a wide variety of other configurations are contemplated depending on each user / operator's specific motion configuration, required structural limitations, special needs, and sensitivity of the intended use of the soft robotic tentacle gripper system.

[0097] 4 is a schematic diagram illustrating an embodiment of a robot assembly 400 including incorporating a soft robotic gripper system with elastomeric sensors 409A, 409B on a controllable palm base plate 403 according to some embodiments of the present disclosure. For example, the soft robotic gripper system of the present disclosure is incorporated into a robot assembly, with a tentacle 411 and two elastomeric sensors 409A, 409B positioned on a palm base plate 402. In an application, at least one goal may be to pick up an object 495 from random objects in a bin 496 based on object shape and pose 497. The tentacle gripper may include a 12 degree of freedom (DoF) and a 6DoF robot arm 404 with tentacles 411. 3D sensors 409A, 409B may be positioned on the arm 404 to acquire data of a scene including the object 495, the tentacle 411, the bin 496, and other objects, weather, etc., which may be used to enhance the robotic assembly process or manufacturing process. The tentacle 411 can then pick up the object from the object 495 from the bin 496. Note that the bin 496 can contain different objects that can be of different sizes and shapes as well as different weights. It is contemplated that the robot can be designed to include seven or more joints, meaning that even though the base is fixed and the gripper position is fixed, the robot's joint positions have one or more extra degrees of freedom, i.e., the "null space" of the robot. It is so called because movement in the null space does not move the gripper position at all. This movement in the null space allows the gripper to reach the object 495 while keeping the gripper from touching another object in the workspace that it should not touch or collide with. For example, extending a human arm outward to a point about two feet in front of the human's face at about nose height creates a null space of the elbow moving up and down without shoulder or hand movement by extending over or under an obstacle such as a pipe obstacle. At least one further innovation is that the multi-tentacle gripper includes a null space, allowing the tentacles 411 themselves to move without changing their position or gripping location on the object 495 being gripped.Thus, the multi-tentacled gripper can circumvent or avoid obstacles in the workspace while maintaining a grip on object 495 with tentacles 411.

[0098] Continuing to refer to FIG. 4, in one example, the 3D sensors 409A, 409B may include additional sensors 409C, 409D, where two or more of the sensors 409A-409D use structured light generated by a projector. Two or more of the sensors 409A-409D may be stereo cameras and time-of-flight range sensors. These sensors 409A-409D may acquire 3D scene data, e.g., point clouds. The 3D sensors may be calibrated to the robot arm 404. Thus, a pose of the object 495 may be estimated in the coordinate system of the 3D sensor, which may be transformed into the coordinate system of the robot arm 404, so that the object 495 may be grasped and picked by controlling the robot arm 404 according to this pose via the robot control computer 443. The scene data may be processed by a control module 442, which may execute a determined method performed in the sensor control computer 401. The sensor control computer 401 may include a memory and input / output interfaces known in the art. It is contemplated that sensors 409A-409D may be configured to help identify and determine that operation of tentacle 411 is proceeding as intended. For example, tentacle 411 grasps object 495 when the multi-tentacled gripping system is commanded to do so at a start location located within a bin of objects 496. This is after the multi-tentacled gripping system is commanded to grasp object 495 from bin 496, but before the multi-tentacled gripping system is commanded to move to a final location. Sensors 409A-409D generate data identifying a type of grip, a number of objects being held by tentacle 411, or both, when the centralized control system receives the sensor data. If the type of grip, the single object being grasped by tentacle 411, or both, is determined to correspond to a predetermined criterion for the object's position and / or orientation according to a set of command instructions, then the centralized control system does not generate an alternative set of command instructions.However, if the centralized control system determines that the grip type, the number of objects grasped by the tentacles 411, or both, do not meet predetermined criteria for the position and / or orientation of the object according to the set of command instructions, the centralized control system generates an alternative set of command instructions that moves the tentacles 411 holding the object or objects back to the starting position or to an alternative position.

[0099] Continuing to refer to FIG. 4, the innovative tentacle 411 is particularly customizable for flexible assembly lines due to the combination of highly dexterous manipulation with real-time sensors 409A-409D as described above, and real-time tactile sensing embedded in the tip (see 116 in FIG. 1C) of each tentacle (not shown). For example, upon receiving real-time sensor data, objects of different shapes and orientations typically encountered in flexible assembly lines can be easily moved by the tentacle with high precision. Test results show that the tentacle is operable to grip two or more types of objects in succession during a period of time, grip an M6 socket head bolt with high precision, grip a casting that the same bolt threads into next in succession, and accurately grip an M6 hex head bolt. Meanwhile, a conventional gripper that grips an M6 socket head bolt is typically unable to grip a casting that the same bolt threads into, nor is it able to accurately grip an M6 hex head bolt.

[0100] At least one novel factor that distinguishes it over conventional robot assemblies is that the tip (see 116 in FIG. 1C) may include an embedded tactile sensor that, in combination with the above-mentioned sensors 409A-409D, provides tactile sensing. When combined, the real-time sensor data returned to a centralized control system (see 300G in FIG. 3G) is processed and executed in a fraction of a second during operation, i.e., within a flexible assembly line. Furthermore, if an error or gripping problem is detected in picking up an object by the tentacle, the centralized control system (see 300G in FIG. 3G) can immediately correct the problem in real time, saving valuable operating time and production costs. This novel feature is particularly important in light of the increasing investments that today's manufacturing facilities are making in incorporating robots with improved capabilities to operate in flexible assembly lines. This is stated in McKinsey & Company's Industrial Robotics Report on insights into the sector's further growth dynamics, and in its July 2019 issue of Advanced Industries, it states that "Increased investments across the industry are driven primarily by the goal of reducing production costs. Investments are also motivated by the need to increase production flexibility and improve robot capabilities (see Evidence 5)" (see https: / / www.mckinsey.com / ~ / media / McKinsey / Industries / Advanced%20Electronics / Our%20Insights / Growth%20dynamics%20in%20industrial%20robotics / Industrial-robotics-Insights-into-the-sectors-future-growth-dynamics.ashx). On the other hand, most conventional robot grippers have little or no tactile sensing, limiting the use of grippers to flexible assembly lines. At least one reason is that while these conventional specialized gripper finger designs are configured for high precision part gripping and grip verification, these conventional devices are not operable to grip more than one type of object.

[0101] 4, while one set of tentacles is shown with supporting motion components, in accordance with an embodiment of the present disclosure, interchangeable sets of tentacles are contemplated depending on a user's specific needs. Different types of sets of tentacles can have different stiffness, flexibility, different exterior surface materials, different tentacle lengths, etc., all of which are described herein in accordance with aspects of the present disclosure.

[0102] Another aspect of the systems and components of the present embodiment is that the software or hardware or some combination thereof can be realized via localized storage or through a cloud-based system. According to an aspect of some embodiments, the tentacles, particularly those configured with embedded sensors, introduce task flexibility, allowing the gripper to continue to be used even if there are changes, predicted or unpredicted, from the original planned task. Similarly, conventional robots have a poor ability to determine whether the actual assembly operation was successful, i.e., whether the gripper's grasp was "good grip", "missing object", "too tight grip", "a little loose grip", "wrong grip", etc. To overcome these above problems of conventional robots, some embodiments of the present disclosure are configured with a mechanically robust generic gripper tip 1216 of Figures 12A-12C, which provides a full set of forces in pressure, translation, rotation, shear, etc., as described above, which can be used for flexible assembly lines, as a non-limiting example.

[0103] 5A-5D are photographs showing types of multi-tentacled grips or grasps for different molded objects that may be used in industrial settings, such as manufacturing for grouping (MFG) packaging applications for small parts, or supply chain applications for bin picking, item sorting, according to some embodiments of the present disclosure.

[0104] What we learned from the experiments is that the combination of a rigid cable guide outer ring, an elastomeric core tube, a stack of elliptical blue spacers, and a steel cable creates a highly anisotropic mechanical metamaterial. Under tension, it is highly inelastic due to the steel cable, and under compression, it behaves unconventionally, i.e. it does not compress axially, nor does it undergo Euler buckling of long columns (which typically creates one sharp fold or kink), but instead bends into essentially a circular arc and recovers perfectly even when bent 180 degrees. Under shear, without cable tension, the tentacles would flex significantly under their own weight, but the tentacle sections themselves would endure a second, higher order curvature ("S" curves and other curves with more inflection points). From a different perspective, the elastomeric gripper is an analog computer that finds the minimum elastomeric energy configuration given the servo cable settings and the boundary conditions of the object being gripped. This perspective led directly to improving the gripper's control algorithm. It was also recognized that the experimental performance results show that the gripper system provides superior grip strength and adaptability through the use of coordinated central control and mechanical metamaterials. Other experimental performance results showed high tensile and compressive strength while remaining flexible in the lateral direction. Additionally, the central controller provides high strength coordinated tentacle gripping, and when compared to a similar conventional gripper of the same size and configuration, the performance results of the gripper of the present disclosure were found to be stronger than a conventional parallel gripper configured with traction rubber gripping jaws and superior in terms of grip type and number for gripping objects much smaller than the minimum tentacle bending radius. For example, FIG. 5A shows multiple tentacles gripping a pencil 510. FIG. 5B shows multiple tentacles gripping a bolt 514. FIG. 7C shows multiple tentacles gripping a car air filter 518 using a first gripping grip. FIG. 5D shows multiple tentacles gripping another car air filter 519 using a second gripping grip.

[0105] FIG. 6A is a photograph showing another type of multiple tentacle grip or grasp for a heavy object, such as a slug hammer 610, according to some embodiments of the present disclosure.

[0106] 6B is a schematic diagram showing a table of grip strength test results for a test gripper having three tentacles and two sections per tentacle, with fully actuated (12 DoF) per tentacle section, + / -120° bend, and a minimum internal bend radius of 30 mm, and the grip strength tested includes grip load capacity, initial and maximum current draw, and failure modes determined for several grips. For example, a multi-tentacle gripper may include many types of grips for the same multi-tentacle, including (a) distal wrap, (b) proximal wrap-around wrap, (c) reverse distal wrap, (d) extended wrap from inside, (e) extended pinch from inside, (f) distal pinch (distal pinch from outside), and (g) proximal pinch. Additionally, the multi-tentacled gripper can grasp very small objects such as a 6 millimeter (M6) sized bolt (or bolts less than ¼ inch sized) to very large objects such as a Shop-Vac industrial vacuum cleaner filter (or objects 150 mm or approximately 6 inches sized) with an encircling grip.

[0107] The gripping load capacity, initial and maximum current draw, and failure modes determined for several grips are summarized in Table 1. After gripping a test object with the tentacles in each type of grip, tests were performed by pulling the test object straight out (axially) or straight down (radially) from the gripper via a calibrated force scale. Excluding the distal pinch, pull-out strengths ranged from 4 to 18 kg and 36 to 160 N (8 to 36 lb). For comparison, a "typical" parallelogram gripping robot gripper with friction rubber jaws, when actuated with two of the same type of HS-805BB servos, achieves only about 15% to 25% of this gripping strength, i.e., 1 to 3 kg lift and 10 to 30 N (2 to 6 lb) axial pull-out strength for similar test objects. Note that some high-performance grips, such as proximal wrap-around wrap, reverse distal wrap, and reverse extension wrap from the inside, require coordinated central control and "unconventional" positioning of the tentacles. Essentially, some tentacles take on a weaker grip to obtain a stronger grip for the tentacle array, which involves diagonally bracing one tentacle against another. These cases illustrate that the optimal grip of a local configuration is not the optimal grip of the whole, and that centralized (rather than distributed) control is necessary. Note that some gripping modes expected to be very strong (such as the full wrap of a boa constrictor) are not possible with only two sections of tentacles with a minimum radius of 30 mm and a bend of + / - 120° per section (such as wrapping around a 10.3 mm diameter test object), and therefore the listed grip strengths should be considered lower limits.

[0108] 7 is a schematic diagram illustrating a robotic system including incorporating a soft robotic gripper system with elastomeric sensors on a multi-tentacled or palmar base plate and a human worker to perform tasks collaboratively as a team in an assembly line of a discrete manufacturing process according to some embodiments of the present disclosure. For example, the soft robotic tentacle system can be incorporated into an assembly line including multiple stations, some of which include at least one task completed by at least one soft robotic tentacle system and at least one task completed by a human or at least one task completed by a collaborative combination of human and soft robotic tentacle system.

[0109] Continuing to refer to FIG. 7, the robot system 790 includes a controller 791 and a robot state detector, e.g., a position encoder 793, which may generate a robot state signal 792. The robot system 790 may also include an object state detector, e.g., cameras 794A, 794B, which may generate an object state signal for an object 795 manipulated by the robot system 790 in a conveyor 712 of a workspace or worktable 711. For example, when measuring grip quality during a predefined test period, a test central control system receives test sensor data, which determines the quality of the gripped object by determining the state of both the tentacle and the object in real time. For example, real-time test sensor data is generated, which may include position and velocity data of points along the tentacle, i.e., points such as joints, which may be used to determine state feedback, which may be used to compare against a predefined command set or to incorporate with other functions related to optimizing the grip, i.e., friction forces in a particular direction, stabilizing control theory, and calculating torques. It should be noted that moving the bag of objects may involve further consideration or calculation due to the bag's dimensions, size, weight, etc., and the tentacles gripping a portion of the bag while carrying it from a start position to a final position in a particular environment. After a number of test periods have been completed, at some point thereafter, a pre-determined grip threshold may be obtained based on the particular test parameters, i.e., object type, object features and characteristics, etc.

[0110] The robot system 790 assists at least one human worker 713 to complete at least one task at the worktable 711, and the workspace or conveyor can move forward and backward to assist the robot or human worker to complete the task. Note that these components 711-712 and 790-794 are depicted here as examples, but may differ for different applications, as the embodiments of the present disclosure are robust to a variety of applications. Additionally, robot motion data can optionally be wirelessly transmitted to and received from the robot learning process 701 depending on the user's specific interests.

[0111] Additionally, the human worker 713 may have sensors to collect data, including wrist device 704, movement monitor 702, on-body movement sensors 706A, 706B, time device 708, and environmental sensor 709. Data from these devices facilitates the learning process for a model specific to the human worker. All these sensors together are an example of a human state detector. The human state is considered along with the state of the manipulated object 795 and the state of the robotic system 790 to learn the control policy of the robot.

[0112] Tentacles designed for experimentation The experimental gripper achieved a kinematically predictable and controllable robot arm, including the ability to wrap around and grasp like an elephant's trunk, by including a tension spring and a constant-length clevis-jointed robot arm with four independent 2DoF sections. But unfortunately, this single-tentacle approach / design was unable to grasp an object with a diameter smaller than the diameter of the tentacle itself, even when driven to a maximum curvature where the inner diameter is equal to zero. That is, the gripper was unable to lift a bolt from the surface of the workspace or grasp a spherical object smaller than the diameter of the tentacle (all such grasps below 270 degrees may prove to be kinematically unstable grasps). Another experimental design of an automated soft robot and biomimetic grasper considered for grasping was configured to grasp based on object recognition, but the actual gripping elements had 1DoF robotic pliers or 1DoF pneumatic soft fingers, which were later found not to be tentacles with a high enough DoF. Thus, what was found was that a higher DoF was needed for the embodiments of the present disclosure.

[0113] In another experimental gripper, a three 2DoF section biomimetic air-driven tentacle robot arm was constructed terminating in a 1DoF gripping tentacle with two rows of vacuum suction to improve grip given a single tentacle. However, it was found that while these vacuum suctions could grip some larger objects, they still had limitations like an elephant's trunk and could not grip objects smaller than the nose diameter. Therefore, upon further testing, this approach could not meet the performance requirements set for the operational performance thresholds of some of the embodiments of this disclosure and was not tested further. Several other experimental devices were analyzed and tested using a gripper design with multi-fingered single-linked rod-like fingers, which, although highly underactuated, prevented sagging or tension exchange between the fingers by placing separate per-finger compliance elements in parallel (rather than a more serial configuration), thereby preventing inter-finger crosstalk and the creation of a more stable grip. Therefore, based on these results, such experimental designs were not developed for this disclosure.

[0114] Other experimental grippers included the design of a series of cable-driven continuous tentacle robot arms with a coiled compression spring core (pseudo leak detector) that provides 3DoF to allow bending in two directions and length changes. However, what was found from these experimental grippers was that the continuous uniform deformations provided by the tentacles with 3 degrees of freedom (DoF) had a relatively small set of grips (encircling and pinching grips) that were much more similar to traditional 1DoF gripper systems, and far less flexible than 12DoF systems. For example, 3DoF multi-tentacle gripper systems cannot perform self-bracing grips such as recurving distal wrap (see Fig. 6B) or reverse extension wrap from the inside (see Fig. 6B).

[0115] Flexibility of tentacle components based on material type A further finding from the experiments is that the amount or degree of flexibility of the tentacle components (i.e., tube sections, tube connectors, rings and spacers) can affect the performance of the tentacle cable transmission mechanism. High flexibility of the tentacle components appeared to reduce bendability, applied gripping force, etc., among other aspects, with respect to operational performance during tentacle operation. In the experiments, several test embodiments were tested with an upper tube section (upper tube, ring with spacer) and a tentacle tube connector that was more flexible than the lower tube (lower tube, ring with spacer). Conversely, one set of test embodiments included a lower tube section and a tentacle tube connector that was more flexible than the upper tube section. Another set of test embodiments included high flexibility for both the lower and upper tube sections and the tube connector. To identify optimal tentacle performance based on a predetermined optimal tentacle performance threshold determined by the tentacle designers of the present disclosure, the materials of each component within the tube section (the tube itself, rings and spacers) were substituted with different flexibility and tube connector materials. Some test experiments included various stiffnesses and flexibilities of the ring versus the spacer in the guide ring design. Some test results showed that increasing the stiffness of the ring material and decreasing the stiffness of the spacer material resulted in improved performance. Additionally, some test results showed that decreasing the flexibility of the ring material and increasing the flexibility of the spacer material resulted in improved performance. As shown in Figure 2D-Figure 2I, tentacle flexibility based on the guide ring geometry can affect performance, and in the test experiments, tentacle performance was further optimized by varying the guide ring geometry to obtain various flexibilities of the tentacles.

[0116] 8A illustrates an embodiment of tentacles attached to a "teach glove" worn by a human being that are used to create a set of motion instructions specific to each pose of an object to move the object, which can be stored in a motion instruction database, where each object can include multiple poses, according to some embodiments of the present disclosure. For example, tentacles 821, 822, 823 attached to a "teach glove" worn by a human being can be used to create a set of motion instructions specific to each predetermined motion action for each object of a number of objects, which can be stored in a motion instruction database, according to some embodiments of the present disclosure. Some benefits / advantages of the multiple tentacles 821, 822, 823 used in a "teach glove" worn by a human being 810 are that the movements of the multiple tentacles 821, 822, 823 of the human wearing the "teach glove" 810 can be recorded along with the appropriate force tensors to be applied during subsequent robot assembly operations.

[0117] 8A , optionally, the robotic system of the present disclosure can be used with other robotic devices over a network to accomplish a variety of collaborative human-like tasks that can be stored in a motion instruction database and later accessed and utilized. For example, the other robotic devices can include devices within which the robotic system of the present disclosure can be implemented to work in unison to accomplish a task (see robotic assembly device of FIG. 4 of the present disclosure having a robot control computer 403, a robot control module 402, and a robot arm 404 in cooperation with multiple tentacles 408).

[0118] As described above, the teach glove 810 is worn by the human operator 301 and can provide control by manipulating a set of sensors 821, 822, 823 attached to the teach glove 801. The movements of the human operator 801 through the teach glove 810 are sensed to provide a set of signals, which are converted into motion sequences or motion instructions to complete a task. A robotic system of the present disclosure in a network of robotic devices can accomplish a portion of a human-like task in a particular environment, while the remaining robotic devices in the network of robotic devices may accomplish the remaining portion of the human-like task in that particular environment. In such a configuration, the motion sequences or motion instructions can be converted into a set of motion instructions stored in a library or motion instruction database. The robotic system of the present disclosure can then access the stored set of motion instructions or programs to execute its motions as if they were consistent with the motions of the operator, and when working with other robotic systems, the robotic system can work in coordination with the other robotic devices to complete a human-like task.

[0119] For example, the centralized robot control system 300G of FIG. 3G may include multiple control sensors 821, 822, 823 of FIG. 8A to detect motion from an operator's hand 801 wearing a teach glove 810. When the operator's hand 801 moves, the control sensors 821, 822, 823 detect the motion and generate output signals representative of the motion. The control sensors 821, 822, 823 may be, for example, accelerometers, or digital positioning devices that provide three-dimensional coordinates for the sensors. As another example, the sensors 821, 822, 823 may measure angular rotation or pressure or force of the joints of the operator 801. The multiple control sensors 821, 822, 823 are attached to the teach glove 810 worn by the operator 801. For example, a radar or lidar, or other 3D depth sensing device, may be positioned several feet away from the operator and pointed toward the operator to detect some of the motion. Various combinations of different types of sensors may be used to detect the motion of the operator 801. A centralized robotic control system 300G of FIG. 3G is communicatively coupled to multiple control sensors 821, 822, 823 and can translate movements sensed by the multiple control sensors 821, 822, 823 into a collective set of commands that actuate multiple tentacles of a robotic system of the present disclosure.

[0120] FIG. 8B is a block diagram illustrating several method steps for creating a set of motion instructions for each pose of an object using the human-worn "teach glove" of FIG. 8A, each pose having an associated set of motion instructions, according to some embodiments of the present disclosure. Step 831 of FIG. 8B includes initializing the robotic system components to establish a robotic system network. Step 833 of FIG. 8B includes, for each movement sequence for transporting an object, detecting human-like motions directly from an operator wearing the glove for the sequence of movements to transport the object, or human-like motions directed by an operator, such as manual manipulation of a control device by an operator to manipulate a tentacle, including the sequence of movements to transport the object. Step 835 of FIG. 8B includes generating signals representing the human-like motions. Step 837 of FIG. 8B includes translating the generated signals into command sequences to drive a network of robotic systems to transport the object. Step 839 of FIG. 8B includes storing each command sequence that drives the network of robotic systems to transport an object in an object movement library, such that each object movement is associated with an object having a particular shape, size, etc.

[0121] Optionally, some other steps of using the teach grab method to sense a movement from an operator wearing the teach grab may include generating one or more signals representative of at least a portion of the movement. Converting the one or more signals into a collective set of commands or motion instructions that actuate the tentacles or actuate the tentacles and other devices in the network of robotic devices, the collective set of commands or motion instructions being functionally equivalent to a motion. Communicating the collective set of instructions or program to a centralized control system and to a network of robotic devices, such as a robotic assembly device, that can be configured to operate with the robotic system. It is contemplated that when communicating the collective set of commands, the communication can be a respective subset of commands from the collective set of commands to each robotic device in the network of robotic devices, i.e., the robotic assembly device. If desired by specific user requirements, the network of robotic devices may have a master robotic device and one or more slave robotic devices, where the master robotic device can receive the collective set of commands or motion instructions and then distribute the collective set of commands or motion instructions to the network of robotic devices, i.e., the robotic assembly device (see FIG. 3A).

[0122] FIG. 9 is a block diagram illustrating some components of another centralized controller system used to execute some methods related to a soft robotic gripper system, according to some embodiments of the present disclosure. For example, the centralized controller system 900 may include a hardware processor 971 in communication with one or more sensors 972 that collect data, including data of the environment 973. The sensor data may include data about robots, vehicles, machines, etc. and similar types of data, or data related to trajectory optimization techniques for dynamic systems with nonlinear constraints. Additionally, the sensor 972 may convert video or camera inputs into signal data. The hardware processor 971 may communicate with a computer storage memory, or memory 979, which includes stored data, including algorithms, instructions and other data that may be implemented by the hardware processor 971. For example, what is stored in the memory may include a multi-link dynamic model, a nonlinear optimization program, and an objective function.

[0123] Continuing with reference to FIG. 9, sensor data may be stored in memory 979, which may be obtained from at least one sensor in the environment via a wired or wireless connection, and the sensor data is stored in the memory via a transceiver configured to output data and receive data, and after each time interval, the sensor data is updated by the at least one sensor. The memory may also store robot, vehicle and machine motion and dynamic data, which may include robot arm information including a model of the robot arm that allows for simulation of the robot arm's movements, along with motion control inputs such as torques applied by motors via robot drives to move each joint of the robot arm. The hardware processor 971 may be connected to a control module 978, which is connected to a robot control computer 980. The robot control computer 980 may communicate back to the control module via 979. The robot control computer 980 is connected to a soft robot gripper system 982, which may communicate with and receive information from the robot control computer 980, and may optionally be directly connected back to the robot control computer 980.

[0124] Optionally, the hardware processor 971 can be connected to a network 977, which communicates with a data source 998, a computer device 984, a mobile telephone device 985, and a storage device 986. Optionally, the hardware processor 971 can also be connected via the network 977 to a network-enabled server (not shown) which is connected to a client device (not shown). The hardware processor 971 can optionally be connected to an external memory device 991, a transmitter 992, and a controller 994. A transceiver can be connected to the hardware processor 971 via an expansion bus interface 995. Some types of output received by the transceiver 996 may relate to a user's intent to receive a response to a request (received by the transceiver 996) for a calculated trajectory along a route (resulting from the method of the present disclosure), which can be displayed on one or more display devices of the user, such as a monitor or screen, and / or can be input to other computer-related devices for further analysis, etc.

[0125] Continuing with reference to FIG. 9, it is contemplated that the hardware processor 971 may include two or more hardware processors as needed for a particular application, and the processor may be either internal or external. Of course, other components may be incorporated depending on the specific operational needs of the user / operator. The network 977 may include, by way of non-limiting example, one or more local area networks (LANs) and / or wide area networks (WANs). The networking environment may be similar to an enterprise-wide computer network, an intranet, and the Internet. It is contemplated that there may be any number of client devices, storage components, and data sources used within the system of the present disclosure. Each may include a single device, or multiple devices working together in a distributed environment. Additionally, the data source 983 may include data resources for training the network. For example, in one embodiment, training data may be stored in the storage 986. The training data may also include other environmental signals. The data source 983 may also include data resources for training the network. The data provided by the data source 983 may include other data, such as other sensor-related data corresponding to the environment, the robotic arm, and the like.

[0126] Continuing to refer to FIG. 9, some data in data sources 983 may be provided by one or more feedback loops. Other examples of data sources may include a variety of sources including, by way of example and not limitation, streaming video, web queries, mobile device camera or other information, webcam feeds, smart glasses and smartwatch feeds, customer care systems, security camera feeds, web documents, catalogs, user feeds, gaming console and user interactions such as SMS logs, instant messaging logs, spoken transcripts, voice commands or data (e.g., data may include types of data such as robots, vehicles, environments, etc., or data related to trajectory optimization techniques for dynamic systems with nonlinear constraints). The particular data source 983 used may be determined based on the application, including whether the data is a particular class of data (e.g., data related to thresholds, performance, safety, etc., data related to robots, vehicles, environments, etc., types of data or data related to trajectory optimization techniques for dynamic systems with nonlinear constraints) or is general in nature (not class specific).

[0127] The third party devices 984, 985 may include any type of computing device, including a computer device 984 or a mobile device 985. It is contemplated that the user device may be embodied as a personal digital assistant (PDA), a mobile device, e.g., a smartphone, a smart watch, smart glasses (or other wearable smart device), an augmented reality headset, a virtual reality headset. Additionally, the user device may be a laptop, e.g., a tablet, a remote control, an entertainment system, a vehicle computer system, an embedded system controller, an appliance, a home computer system, a security system, a home electronic device, or other similar electronic device. In one embodiment, the client device may receive input data, such as voice and information usable by the methods and systems of the present disclosure. For example, the third party device may be data, i.e., data related to a robot, a vehicle, an environment, or similar types of data, or data related to trajectory optimization techniques for dynamic systems with nonlinear constraints, a microphone or line input terminal for receiving voice information, a camera for receiving video or image information, or a communication component (e.g., Wi-Fi capability) for receiving such information from another source, such as the Internet or a data source 3.

[0128] 9, with respect to storage 986, storage 986 may store information including data, computer instructions (e.g., software program instructions, routines, or services), and / or models used in embodiments of the technology described herein. For example, storage 986 may store data from one or more data sources 983, one or more deep neural network models, information for generating and training the deep neural network models, and computer-usable information output by the one or more deep neural network models.

[0129] 10 is a schematic diagram illustrating a centralized controller system that may be used to execute some methods related to the soft robotic gripper system according to some embodiments of the present disclosure. For example, an alternative centralized controller system may be a computing device representing various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers.

[0130] The centralized controller system 1000 may include a power supply 1008, a processor 1009, a memory 1010, and a storage device 1011, all of which are connected to a bus 1050. Additionally, a high-speed interface 1012, a low-speed interface 1013, a high-speed expansion port 1014, and a low-speed connection port 1015 may be connected to the bus 1050. Also, a low-speed expansion port 1016 connects to the bus 1050. Various component configurations are contemplated that may be mounted on a common motherboard 1030, as a non-limiting example, depending on the particular application. Additionally, an input interface 1017 may be connected to an external receiver 1006 and an output interface 1018 via the bus 1150. A receiver 1019 may be connected to an external transmitter 1007 and a transmitter 1010 via the bus 1050. An external memory 1004, an external sensor 1003, a machine 1002, and an environment 1001 may also be connected to the bus 1050. Additionally, one or more external input / output devices 1005 may be connected to the bus 1050. A network interface controller (NIC) 1021 may be adapted to connect to a network 1022 through the bus 1050 to, among other things, render data or other data on a third party display device, a third party imaging device, and / or a third party printing device external to the computing device 1000.

[0131] It is intended that the memory 1010 may store instructions executable by the centralized controller system 1000, historical data, and any data available to the methods and systems of the present disclosure. The memory 1010 may include random access memory (RAM), read only memory (ROM), flash memory, or other suitable memory systems. The memory 1010 may be one or more volatile memory units and / or one or more non-volatile memory units. The memory 1010 may be another form of computer readable medium, such as a magnetic or optical disk.

[0132] Continuing with reference to FIG. 10, the storage device 1011 may be adapted to store supplemental data and / or software modules used by the computing device 1000. For example, the storage device 1011 may store historical data and other related data as described above with respect to this disclosure. Additionally or alternatively, the storage device 1011 may store historical data similar to that described above with respect to this disclosure. The storage device 1011 may include a hard drive, an optical drive, a thumb drive, an array of drives, or any combination thereof. Additionally, the storage device 1011 may include a computer readable medium such as a floppy disk device, a hard disk device, an optical disk device, or an array of devices including a tape device, a flash memory or other similar solid state memory device, or a storage area network or other configuration of devices. Instructions may be stored on the information carrier, which, when executed by one or more processing units (e.g., the processor 1009), perform one or more methods, such as those described above. The system can optionally be linked through a bus 1050 to a display or user interface (HMI) 1023 adapted to connect the system to a display device 1025 and a keyboard 1024, which may include a computer monitor, a camera, a television, a projector, or a mobile device, among others.

[0133] Continuing to refer to FIG. 10, the centralized controller system 1000 can include a user input interface 1017 adapted to a printer interface (not shown) that can also be adapted to connect through a bus 1050 to a printing device (not shown), which can include a liquid inkjet printer, a solid ink printer, a large scale commercial printer, a thermal printer, a UV printer, or a dye-sublimation printer, among others. The high-speed interface 1012 manages the bandwidth-intensive operations of the computing device 1000, while the low-speed interface 1013 manages the less bandwidth-intensive operations. Such an allocation of functions is merely exemplary. In some implementations, the high-speed interface 1012 can be coupled (e.g., through a graphics processor or accelerator) to the memory 1010, the user interface (HMI) 1023, and to a keyboard 1024 and a display 1025, and can also be coupled via the bus 1050 to a high-speed expansion port 1014 that can accept various expansion cards (not shown). In this implementation, low speed interface 1013 is coupled to storage device 1011 and low speed expansion port 1015 via bus 1050. Low speed expansion port 1015, which may include a variety of communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet), may be coupled, for example through a network adapter, to one or more input / output devices 1005 and other devices such as a keyboard 1024, a pointing device (not shown), a scanner (not shown), or a networking device such as a switch or router.

[0134] 10, the centralized controller system 1000 may be implemented in a number of different forms, as shown. For example, it may be implemented as a standard server 1026, or multiple times in a group of such servers. In addition, it may be implemented in a personal computer, such as a laptop computer 1027. It may also be implemented as part of a rack server system 1028, and an external power supply 1029. Alternatively, components from the computing device 1000 may be combined with other components in a mobile device (not shown), such as a mobile computing device having a different component configuration arranged for the mobile device.

[0135] Embodiment The following description provides only exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. Rather, the following description of exemplary embodiments will provide those skilled in the art with an enabling description for implementing one or more exemplary embodiments. It is contemplated that various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the disclosed subject matter, as set forth in the appended claims.

[0136] Specific details are provided in the following description to provide a thorough understanding of the embodiments. However, those skilled in the art will appreciate that the embodiments may be practiced without these specific details. For example, systems, processes, and other elements in the disclosed subject matter may be shown as components in block diagram form in order to avoid obscuring the embodiments in unnecessary detail. In other instances, well-known processes, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments. Additionally, like reference numbers and names in the various drawings refer to like elements.

[0137] Although the present disclosure has been described with reference to certain preferred embodiments, it must be understood that various other adaptations and modifications can be made within the spirit and scope of the disclosure. It is therefore the aspect of the following claims to cover all such variations and modifications that come within the true spirit and scope of the disclosure.

Claims

1. An apparatus comprising: The device comprises tentacles, each of the tentacles including a lower elastic member and an upper elastic member connected by an elastic connector, the elastic connector including a central through hole and a transmission channel in a central portion, each elastic member includes a guide disk, each of the guide disks includes a ring having a through hole, a spacer having a through hole is located in a donut hole of the ring, the ring and the through hole of the spacer together define a cable path, the lower elastic member is composed of guide rings of different thicknesses, the number of guide rings with a smaller thickness is greater than the total number of guide rings with an intermediate thickness and guide rings with a larger thickness, the lower elastic member is arranged from one end to the other end in the order of the thicker guide ring, the intermediate thickness guide ring, and the thinner guide ring; The device further comprises a cable having a proximal end coupled to an actuator external to the lower elastic member and extending from the actuator through an opening in a controllable palm baseplate located at the proximal end of the lower elastic member.

2. 2. The apparatus of claim 1, wherein each said transmission channel includes a geometric feature having an entrance opening having a cable entrance path tangent to an inner surface of the central through hole and an exit opening having a cable exit path parallel to a central axis of the elastic connector.

3. 3. The apparatus of claim 2, wherein the cable entry path and the cable exit path together minimize an amount of transmission friction and minimize an amount of cable flex fatigue while the cable is moving during operation of the tentacle.

4. 3. The apparatus of claim 2, wherein the geometric characteristics are based on obtaining a measured distance from the guide disk through hole to the transmission channel, and configuring the cable inlet path and the cable outlet path using the measured distance and a pair of circular arcs of constant and equal radius to determine an optimal transmission channel configuration.

5. 5. The apparatus of claim 4, wherein maximizing the radius of the pair of arcs within the measurement distance minimizes the amount of side load on the cable resulting from movement of the cable while simultaneously minimizing the amount of transmission friction and the amount of cable flex fatigue during operation of the tentacles.

6. 3. The apparatus of claim 2, wherein the cable entry path extends from a bottom surface to a top surface of the flexible connector at an acute angle, and the cable exit path extends from the top surface to the bottom surface at an acute angle.

7. 3. The device of claim 2, wherein each inlet opening of the transmission channels is approximately aligned with a lower spacer through hole of a distal lower guide ring and each outlet opening of the transmission channels is approximately aligned with an upper ring through hole of a proximal upper guide ring.

8. The apparatus of claim 1 , wherein one or more of the guide discs include convex bottom and top surfaces and rounded peripheries or convex edges.

9. The apparatus of claim 1 , wherein a material forming the guide ring has a different stiffness for the ring than a material forming the spacer.

10. The device of claim 1 , wherein the palm base plate is controllable along an X-axis corresponding to a left-right direction and a Y-axis corresponding to a front-back direction.

11. 2. The device of claim 1, wherein a set of lower cables extend through the through holes of the lower ring and are coupled to the distal lower guide disk, a set of upper cables extend through the through holes of the lower spacer, through the transmission channel to the through holes of the upper ring and are coupled to the distal upper guide ring, and an end cap is attached to the distal end of the upper elastic member.

12. 2. The apparatus of claim 1, wherein the smaller thickness guide ring optimizes flexibility of the tentacle compared to the larger thickness guide ring providing maximum load capability of the tentacle and the intermediate thickness guide ring providing an equal level of flexibility and load capability of the tentacle.

13. 2. The device of claim 1, wherein the tentacles include three tentacles with 12 independent degrees of actuation freedom, and wherein a tension cable has a bending range of 120° for each of the lower elastic member and the upper elastic member, thereby providing multiple grip modes, the multiple grip modes including a distal pinch grip expanding from the inside, a reverse expanding wrap from the inside, a proximal enveloping wrap, and a reverse distal wrap.

14. 10. The apparatus of claim 1, further comprising a partial grab device fitted to the tentacles and attached to the palm base plate, the partial grab device including reinforcing support rib portions axially aligned with a number of lower member guide rings, and an object contacting surface of the partial grab device including one or more gripping features disposed between each tentacle.

15. A gripper device, the gripper device comprises tentacles, each tentacle including a lower member and an upper member connected by a connector, each member having a guide disk mounted along said members to maintain spacing between said guide disks, each guide disk including a ring having a through hole, a spacer having a through hole located within a donut hole of said ring, the through holes of said ring and said spacer together defining a cable path, said through holes of each ring being equally spaced along a circumference of said ring, said connector including a central through hole and a central transmission channel, each said transmission channel including a geometric feature having an entrance opening having a cable entrance path tangent to an inner surface of said central through hole and an exit opening having a cable exit path parallel to a central axis of said connector; The gripper device further comprises cables having proximal ends coupled to actuators external to the lower member and extending from the actuators through openings in a controllable palm base plate located at the proximal end of the lower member, a pair of lower cables extending through holes in a lower ring to couple to a distal lower guide disk, a pair of upper cables extending through holes in a lower spacer of the lower guide ring, through the transfer channels to holes in an upper ring to couple to a distal upper guide ring, and an end cap attached to a distal end of the upper member; The gripper apparatus further comprises a partial grab device, the partial grab device being attached to the tentacle and the palm base plate, the partial grab device including a reinforcing support rib portion axially aligned with a number of lower member guide rings.

16. 16. The gripper apparatus of claim 15, wherein each inlet opening of the transmission channels is generally aligned with a through hole in a lower spacer of a distal lower guide ring and each outlet opening of the transmission channels is generally aligned with a through hole in an upper ring of a proximal upper guide ring, one or more of the guide disks include convex bottom and top surfaces and rounded peripheral or convex edges, and the flexibility of the material forming the guide disks is different from the flexibility of the material forming the spacers.

17. The gripper device of claim 15 , wherein the partial grabber device is formed from a material having an amount of elasticity, an amount of stiffness, or both selected for moving an object within a working area.

18. 1. A soft robotic gripper system, comprising: the soft robotic gripper system comprises tentacles, each tentacle including a lower member and an upper member connected by a connector, each member having a guide disk mounted along the member to maintain spacing between the guide disks, each guide disk including a ring having a through hole, a spacer having a through hole located within a donut hole of the ring, the ring and the through holes of the spacer together defining a cable path, the connector including a central through hole and a transmission channel in a central portion, each transmission channel including a geometric feature having an entrance opening having a cable entrance path tangent to an inner surface of the central through hole and an exit opening having a cable exit path parallel to a central axis of the connector; The soft robotic gripper system further comprises cables having proximal ends coupled to actuators external to the lower member and extending from the actuators through openings in a controllable palm baseplate located at the proximal end of the lower member, a set of lower cables extending through holes in a lower ring to couple to a distal lower guide disk, a set of upper cables extending through holes in a lower spacer of the lower guide ring, through the transfer channels to holes in an upper ring to couple to a distal upper guide ring, and an end cap attached to a distal end of the upper member; The soft robotic gripper system further comprises a centralized control system, the centralized control system comprising: Receive tentacle data and sensor data from the transceiver; comparing the sensor data with stored object configurations to obtain corresponding stored object configurations and comparing stored commands to obtain a set of stored commands corresponding to the stored object configurations; using the received tentacle data to obtain a corresponding set of stored tentacle actions by comparing stored tentacle actions associated with the stored object configuration, and to obtain a corresponding first set of commands by comparing the stored set of commands, and if the tentacle actions from the tentacle data do not correspond to the stored object configuration, selecting a second set of commands; generating a series of control signals that cause a motor for each one of the tentacles to apply a series of tensions to a transmission system to each tentacle according to an associated one of the first set of commands or the second set of commands to move an object; A soft robotic gripper system configured as described above.

19. 20. The soft robotic gripper system of claim 18, wherein the first set of commands and the second set of commands include a series of predetermined torques, the torques along with corresponding tensions are applied to one or more joints or bending joints and to a plurality of transmission systems of each one of the tentacles, each transmission system having a first end connected to the motor and a second end connected to the joint or bending joint.

20. 20. The soft robotic gripper system of claim 18, wherein the sensor data includes a distal joint sensor located at at least one tip of at least one tentacle, the distal joint sensor including a distributed shape sensor or a linear displacement sensor.

21. 1. A soft robotic gripper system, comprising: the soft robotic gripper system comprises tentacles, each tentacle including a lower member and an upper member connected by a connector, each member having a guide disk mounted along the member to maintain spacing between the guide disks, each guide disk including a ring having a through hole, a spacer having a through hole positioned within a donut hole of the ring, the ring and the through hole of the spacer together defining a cable path; the connector includes a central through hole and a central transmission channel, each of the transmission channels including a geometric feature having an entrance opening with a cable entrance path tangent to an inner surface of the central through hole and an exit opening with a cable exit path parallel to a central axis of the connector; The soft robotic gripper system further comprises cables having proximal ends coupled to actuators external to the lower member and extending from the actuators through openings in a controllable palm baseplate located at the proximal end of the lower member, a set of lower cables extending through holes in a lower ring to couple to a distal lower guide disk, a set of upper cables extending through holes in a lower spacer of the lower guide ring, through the transfer channels to holes in an upper ring to couple to a distal upper guide ring, and an end cap attached to a distal end of the upper member; The soft robotic gripper system further comprises a centralized control system, the centralized control system comprising: Receive tentacle data and sensor data from the transceiver; comparing the sensor data with stored object configurations to obtain corresponding stored object configurations and comparing stored commands to obtain a set of stored commands corresponding to the stored object configurations; the centralized control system is further configured to: use the received tentacle data to obtain a corresponding set of stored tentacle actions by comparing stored tentacle actions associated with the stored object configurations; obtain a corresponding first set of commands by comparing the stored sets of commands; and select a second set of commands if the tentacle actions from the tentacle data do not correspond to the stored object configurations; the first set of commands and the second set of commands include a series of predetermined torques, the torques being applied to one or more joints or bending joints together with corresponding tensions, and being applied to a plurality of transmission systems of each of the tentacles, each of the transmission systems having a first end connected to a motor and a second end connected to the joints or bending joints; and a soft robotic gripper system configured to generate a set of control signals that cause the motor for each of the tentacles to apply a set of tensions to the transmission system to each of the tentacles in accordance with an associated first set of commands or a second set of commands to move an object.

Images