Data

The National Institute of Standards and Technology, Intelligent Systems Division has collected data measuring human subjects, while performing common, simulated industrial manufacturing tasks with and without wearing an exoskeleton. Five tests were completed as part of a research study to develop measurement science towards standard test methods. For simulated industrial manufacturing tasks were performed using a novel, now standardized apparatus, called the Position and Load Test Apparatus for Exoskeletons (PoLoTAE). In addition, a set of novel optical tracking marker artifacts were worn by the subject for synchronous tracking of exoskeleton and human leg position and orientation. The standard test artifacts were intended to address the challenges of measurement uncertainty variation between different marker clusters and marker movement on soft tissue and marker occlusion when using traditional bio-mechanical marker models while wearing an exoskeleton. The PoLoTAE tests simulated generic industrial tasks (load positioning, load alignment, peg-in-hole, applied force). The knee bend tests were performed to synchronously track the exoskeleton and human lower limb position and orientation for analysis such as comparing the exoskeleton fit to the subject’s leg. 

Overall, the tests included 116 subjects of which 68 subjects (59% of total subjects) consented to publication of their raw test data described in this paper.  While some subjects performed more than one test, at least 30 subjects performed each of the five tests totaling 158 tests performed. To date, aggregate data for the load positioning and knee bend tests have been analyzed and are referenced in this paper. Sensor data was collected from each subject,  which included: repetition number, heart rate, videos, skeletal joint pose estimation, and survey data.

This data was developed at the National Institute of Standards and Technology by employees of the Federal Government in the course of their official duties. Pursuant to title 17 Section 105 of the United States Code this data is not subject to copyright protection and is in the public domain. This work is resulting from an experimental system. NIST assumes no responsibility whatsoever for its use by other parties, and makes no guarantees, expressed or implied, about its quality, reliability, or any other characteristic. We would appreciate acknowledgement if the data is used. This data can be redistributed and/or modified freely provided that any derivative works bear some notice that they are derived from it, and any modified versions bear some notice that they have been modified. No approval or endorsement of any commercial product by the National Institute of Standards and Technology is intended or implied. Certain commercial equipment, instruments, or materials may be identified in this data in order to facilitate understanding. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified are necessarily the best available for the purpose.

• Generic Test Recruitment and Procedures
• Position and Load Test Apparatus for Exoskeletons (PoLoTAE)

Test Subject Recruitment
Test subjects were recruited via flyers and internal NIST website posts based on the following criteria: 1) physically fit to perform the tasks (i.e., can wear an exoskeleton that weighs approx. 30.3 Lbs (13.7 kg), can perform knee bends, position tools, and apply forces 30 times twice (60 times total) using tools (up to the approx. weight of 2 gallons of milk); 2) are at least 18 years old; and 3) fit within the exoskeleton manufacturers specification for height of 1.52 m to 1.85 m (5 ft 0 in to 6 ft 1 in) and weight of 49 kg to 102 kg (108 Lbs to 225 Lbs).  For the Load Positioning, Knee Bend, Peg-In-Hole, and Load Alignment Tests, recruited subjects were federal employees.  Protocol was later approved to recruit non-federal employees for the Applied Force and for future tests.

Generic Test Procedures

All subjects signed a consent form allowing the research team to collect and store data from the test. Also, the data was allowed to be analyzed and published in the aggregate (i.e., combined subjects for the same test). To leverage the exoskeleton community of researchers to advance the test, measurement and analysis methods  an addendum was sent to all subjects requesting their raw test data to be published.  The data presented here are provided for publication, and therefore consented by the subject. The data includes 68 subjects of the total 116 subjects (59% of total subjects).

Each subject was provided a subject number prior to testing where their name was not used beyond recruitment (i.e., during testing and for data storage and analysis). The test to be performed was explained in detail to the subject by the research staff.  A training video was  reviewed by each subject prior to testing. The video displayed the test procedure, the sensors used (e.g., heart rate monitor), the type of test to be performed, and an example survey to be completed about the test.  If the subject agreed to  perform the test, they signed the consent form accordingly. Each test subject also self declared any discomfort such as soreness before the test.
 For the load positioning test, each subject performed the simulated task without the exoskeleton prior to performing the simulated task with the exoskeleton. For subsequent tests, the baseline and exoskeleton tests were alternated per subject (i.e., begin one subject with the baseline test, the next subject began with wearing an exoskeleton, etc.). For exoskeleton tests, the research team assisted with donning and doffing of the exoskeleton.  

Terminology to clearly describe the tests are as follows:

• test, n—a collection of trials or task repetitions.
• trial, n—a collection of repetitions for a single test set (a single type of simulated industrial task).
• repetition, n—performance of a task.

An example that uses the above terminology to describe a typical test performed in this paper is: “each subject completed a ‘test’ including a baseline ‘trial’ of 30 ‘repetitions’ and exoskeleton(s) ‘trial(s)’ of 30 ‘repetitions’.”

The subject was seated to allow a resting heart rate to be recorded on the test form by the researcher. Optical  heart rate monitors were attached to the subject’s wrists and video recording began. The subject stood at a start line and when all sensors started recording data , the subject began the test. Upon completion of each set of repetitions, an electrical, handgrip, heart rate monitor was also used to measure     the subject’s heart rate.  As optical heart rate monitors sometimes varied greatly for subjects, a comparison heart rate results was performed where the optical heart rate monitor reading closest to the electronic heart rate monitor reading was selected as the documented heart rate. However, video recording of heart rate monitors throughout the test, as well as direct heart rate monitor test data was recorded for both monitors and is provided.  Quad video from each test subject performing a task test was recorded and is provided.  The video was split and added in a quad video format to include rear right and left  views of the subject along with sensor data.

To date, five of the seven originally planned tests were completed including, four different task tests and one fit test, where one of the tests was setup and performed at a time.  When at least 30 subjects were tested , the test was considered completed and another test was configured.  Each test took approximately 30 min to 45 min for the subject to complete. For nearly all subjects who performed the fit test (i.e., 1b) the task test 1a was initially performed and the subject walked within the exoskeleton approximately 100 steps to ensure that the exoskeleton had settled properly onto the test subject.

[Top]

NIST researchers designed and developed  the Position and Load Test Apparatus for Exoskeletons (PoLoTAE) to begin standardization of industrial tasks when subjects wear exoskeletons. The PoLoTAE is self-standing and measures 3 m wide x 2.4 m high x 1.2 m deep (10 ft wide x 8 ft high x 4 ft deep). The Load Artifact was used for Load Positioning and Load Alignment task tests.  The load artifact measured  11.2 cm long  x 4.9 cm wide   x 2.1 cm high (28.5 in long x 12.5 in wide x 5.25 in high) and weighed 6.8 kg (15 Lbs). A typical battery powered drill was used as the Peg-In-Hole tool for the Peg-in-Hole task test and a typical hand-held grinder was used as the Grinder tool for the Applied Force task test.

[Top]