Single-Action-Lock Structural Space Joint

Assemblers are a team of modular robots that work together to build things. Each Assembler is a stack of one or more Stewart platforms, or hexapods, made up of two plates connected by six linear actuators for movement, enabling a full six-degree-of-freedom (DOF) pose of the top plate relative to the bottom plate (see figure to the right). An end effector on each Assembler enables gripping, lifting, and welding/joining. The Assemblers system architecture features novel control algorithms and software, sensors, and communicator technology that coordinate operations of Assembler teams. The control system includes an important module for task management that estimates how many robots are needed, the optimal number of hexapods in each Assembler, and the estimated voltage needed. There are also modules for trajectory generation, joint control, sensor fusion, and fault detection. The novel control system directs the Assembler operations for high accuracy and precision, yet there is built-in dynamic resilience to failure. For example, if a single hexapod on an Assembler fails, the system deems it “rigid” in its last pose and redistributes the work to the other Assemblers. The image below shows a storyboard of operations for how Assemblers might build a solar array. NASA has developed a hardware demo with communications between subsystems, backed up by detailed simulations of the kinematics and actuator dynamics.

To enable the goal of autonomous assembly of high performing structures, a robot system must be able to travel across a lattice structure in all dimensions, transport and align a unit cell module to the correct location and fasten the module to the existing structure. In this system, a team of multiple mobile bipedal robots work together to carry, transfer, and place 3D-lattice modules (e.g., cuboctahedron voxels) to form a 3D lattice structure. The team of mobile bipedal robots autonomously provide transportation, placement, unpacking, and assembly of voxel modules into functional structures and systems. As the team of mobile bipedal robots live and locomote on the 3D-lattice structure, they monitor health and performance, enabling repair and reconfiguration when needed. The mobile bipedal robots work together in different roles, for example, one as a cargo transport robot and the other as a crane robot. The cargo transport robot and the crane robot work together to move the voxels from one location to another. Each robot includes at least one electronic control module that receives commands from another robot or a central control system. A central control system implements a plan to control the motion sequences of the robots to maximize efficiency and to optimize the work required to completely assemble a structure. The plan is pre-computed or computed during implementation by the central control system or the robots themselves, according to algorithms that utilize the regularity of the lattice structure to simplify path planning, align robotic motions with minimal feedback, and minimize the number of the degrees of freedom required for the robots to locomote across and throughout the 3D-lattice structure and perform structural assembly.

NASA, GM, and Oceaneering approached the development of R2 from a dual use environment for both space and terrestrial application. NASA needed an astronaut assistant able to function in space and GM was looking for a robot that could function in an industrial setting. With this in mind, R2 was made with many capabilities that offer an enormous advantage in industrial environments. For example, the robot has the ability to retool and vary its tasks. Rather than a product moving from station to station on a conveyor with dozens of specialized robots performing unique tasks, R2 can handle several assembly steps at a single station, thereby reducing manufacturing floor space requirements and the need for multiple robots for the same activities. The robot can also be used in scenarios where dangerous chemicals, biological, or even nuclear materials are part of the manufacturing process. R2 uses stereovision to locate human teammates or tools and a navigation system. The robot was also designed with special torsional springs and position feedback to control fine motor movements in the hands and arms. R2's hands and arms sense weight and pressure and stop when they come in contact with someone or something. These force sensing capabilities make R2 safe to work side-by-side with people on an assembly line, assisting them in ergonomically challenging tasks or working independently. This NASA Technology is available for your company to license and develop into a commercial product. NASA does not manufacture products for commercial sale.

Originally developed by NASA and GM, the Robo-Glove technology was a spinoff of the Robonaut 2 (R2), the first humanoid robot in space. This wearable device allows the user to tightly grip tools and other items for longer periods of time without experiencing muscle discomfort or strain. An astronaut working in a pressurized suit outside the space station or an assembly operator in a factory might need to use 15 to 20 lbs of force to hold a tool during an operation. Use of the Robo-Glove, however, would potentially reduce the applied force to only 5 to 10 lbs. The Robo-Glove is a self-contained unit, essentially a robot on your hand, with actuators embedded into the glove that provide grasping support to human fingers. The pressure sensors, similar to the sensors that give R2 its sense of touch, are incorporated into the fingertips of the glove to detect when the user is grasping an object. When the user grasps the object, the synthetic tendons automatically retract, pulling the fingers into a gripping position and holding them there until the sensor is released by releasing the object. The current prototype weighs around two pounds, including control electronics and a small display for programming and diagnostics. A lithium-ion battery, such as one for power tools, is used to power the system and is worn separately on the belt.

The Multi-Link Spherical Joint developed at NASA Johnson Space Center provides a substantial improvement over typical joints in which only two linearly actuated links move independently from one another. It was determined that the rotation point of a trussed link needed to be collocated at a shared point in space for maximum articulation. If not allowed separate rotation, the line of action through a universal joint and hinge acts effectively as another linkage. This leads to a much more complex and uncontrollable structure, especially when considering multiple dimensions. Comprising the Multi-Link Spherical Joint, a spherical shell encases the cupped ends of each six possible attachments and allows each of those attachments to be independently controlled and rotated without inhibiting the motion of the others. To do this, each link is precisely limited to 15 degrees of rotation off the link centerline, thus allowing a total of 30 degrees of rotation for each link. The shell-and-cup structure can handle the loads of linear actuators that may be used to control and vary the geometry of a truss system utilizing the new joint technology. The calculated operating load that the truss system must handle can be used to scale the size of the joint, further allowing customization of any potential truss system. Additionally, the incorporated linear actuators can be controlled and powered by wiring routed through the joint without putting undo stress on the wires during operation. Accordingly, this innovative joint technology enables more efficient deployment and precise operation of articulating structures. The Multi-Link Spherical Joint is at technology readiness level (TRL) 4 (component and/or breadboard validation in laboratory environment) and is available for patent licensing. Please note that NASA does not manufacture products itself for commercial sale.