Videos from the Measurement Science for Manufacturing Robotics Program
Agile Robotics for Industrial Automation Competition 2019
This year the Agile Robotics for Industrial Automation Competition (ARIAC) saw a record number of qualifying teams and participants. ARIAC is simulation-based and designed to promote robot agility by utilizing the latest advances in artificial intelligence and robot planning. The objective of the competition is to test the agility of industrial robot systems, with the goal of enabling industrial robots on the shop floors to be more productive, more autonomous, and to require less time from shop floor workers. This video demonstrates how competing teams addressed different challenges during their kit building objectives.
Enabling Agility in Industrial Robotic Applications
In this video, two robots from different companies are performing independent kitting operations. The two robots have fundamentally different underlying programming languages. One robot enters a simulated error state. The second robot is dynamically retasked to complete the first robot’s task by using the NIST-developed Canonical Robot Command Language (CRCL). The robot does so without any human interaction. Once completed, the robot goes back to completing its own task.
Robotic Hand Performance Testing
Next-generation robotic hand technologies will help to reduce the costs of custom robotic tooling through their flexibility and dexterity for handling a large variety of part geometries and manufacturing operations. NIST supports these emerging technologies by providing performance measurement tools that in the short term are used to benchmark research and in the long term will help integrators and end-users select the best robotic hand technology to meet their application needs. Robotic hands have various key characteristics that are fundamental to their real-world performance. This video illustrates several recently developed test methods that measure these traits, including finger strength, grasp strength, slip resistance, touch sensitivity, force tracking, and manipulation. Performing these tests reveal insights into the hand’s strength, lightness of touch, and ability to perform dexterous in-hand manipulation of objects.
Impact of Robotic Hand on Pick-and-Place Performance
NIST is developing metrics and test methods to benchmark the performance of robotic systems when performing manufacturing tasks. Here, the speed and dexterity of pick-and-place robotic systems are assessed using a variant of the Minnesota Dexterity Test, a test designed to measure human capability for relatively simple hand-eye coordination. This video demonstrates a slight modification to the original displacement test, where puck-like objects are displaced one-by-one across the entirety of the board and placement accuracy is quantified via a set of concentric circles that indicate a target zone. With completion time and placement accuracy as the principal measures of performance, a robotic system conducts the test with two different robotic hands to quantify and compare their effects on the overall system performance.
Comparative Peg-in-Hole Testing of Robotic Hand and Gripper
NIST is developing metrics and test methods to benchmark the performance of robotic systems when performing manufacturing tasks. The ability to perform simple insertions is critical for robotic systems in manufacturing. This video reveals a simple peg-in-hole test designed to measure a robotic system’s capability for performing these simple insertions. In this case, the robotic system is outfitted with two different end-effectors – a robotic hand and a robotic gripper – to study the performance of next-generation robotic hand technology versus conventional parallel gripper technologies.
Coordinated assembly in a heterogeneous robotic work cell
Multiple robots are programmed to complete a complex kitting, part transfer, and assembly operation. The work cell consists of five robots from different manufacturers, and are controlled using NIST-proprietary software for coordination such that they all operate in the same coordinate system. Parts are picked from feeders and placed into a kit tray. That kit tray is then transferred to a second work cell via hand-offs between robots. At the second work cell, the parts are then removed from the kit tray and physically mated to form a final assembly. Once the assembly is completed, the part is returned to the kit tray, and returned to the first work cell, where it is removed, and the kit tray is reset. The process then restarts with a new assembly.
Navigation Test Method
ASTMF45.02 developed a test method for A-UGVs to navigate within defined spaces. This video demonstrates an early set of tests using an Automatic-UGV (AGV) to navigate within a defined space. The space was continuously made more narrow to measure when the vehicle deviated from the commanded path and the vehicle stopped when the on board safety sensors detected the defined space barriers. The navigation in defined space standard was modified to include defined space shapes and pass/fail criteria for each set of repetitions with barrier spacing chosen by the requester. [No audio]
Docking Test Method
ASTMF45.02 developed a test method for A-UGVs to dock to fixed equipment. This video demonstrates an early set of tests using a fork-style Automatic-UGV (AGV-automatic guided vehicle) to navigate to an apparatus with targets. The fork tines were raised and lowered to align with the targets on the frame. Also, two points extended from the A-UGV side were aligned with the frame mocking up the scenario of aligning a unit-load vehicle to a tray station. The docking work item is currently being drafted and may differ from the video. [No audio]