CoChat

The DOI matches its Crossref record; 35 of the 39 references it cites were found in a registry or an index.

Human-robot cooperative movement training: Learning a novel sensory motor transformation during walking with robotic assistance-as-needed

Open access
Emken, Jeremy L, Benitez, Raul, Reinkensmeyer, David J · 2007 · Journal of NeuroEngineering and Rehabilitation

Abstract

Background: A prevailing paradigm of physical rehabilitation following neurologic injury is to "assist-as-needed" in completing desired movements. Several research groups are attempting to automate this principle with robotic movement training devices and patient cooperative algorithms that encourage voluntary participation. These attempts are currently not based on computational models of motor learning. Methods: Here we assume that motor recovery from a neurologic injury can be modelled as a process of learning a novel sensory motor transformation, which allows us to study a simplified experimental protocol amenable to mathematical description. Specifically, we use a robotic force field paradigm to impose a virtual impairment on the left leg of unimpaired subjects walking on a treadmill. We then derive an "assist-as-needed" robotic training algorithm to help subjects overcome the virtual impairment and walk normally. The problem is posed as an optimization of performance error and robotic assistance. The optimal robotic movement trainer becomes an error-based controller with a forgetting factor that bounds kinematic errors while systematically reducing its assistance when those errors are small. As humans have a natural range of movement variability, we introduce an error weighting function that causes the robotic trainer to disregard this variability. Results: We experimentally validated the controller with ten unimpaired subjects by demonstrating how it helped the subjects learn the novel sensory motor transformation necessary to counteract the virtual impairment, while also preventing them from experiencing large kinematic errors. The addition of the error weighting function allowed the robot assistance to fade to zero even though the subjects' movements were variable. We also show that in order to assist-as-needed, the robot must relax its assistance at a rate faster than that of the learning human. Conclusion: The assist-as-needed algorithm proposed here can limit error during the learning of a dynamic motor task. The algorithm encourages learning by decreasing its assistance as a function of the ongoing progression of movement error. This type of algorithm is well suited for helping people learn dynamic tasks for which large kinematic errors are dangerous or discouraging, and thus may prove useful for robot-assisted movement training of walking or reaching following neurologic injury.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/1743-0003-4-8

Cites

35 of 35 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Conceptual overview of human-robot cooperative, assist-as-needed gait training. The goal of this type of training is to allow the human to learn to compensate for the gait impairment, but to reduce the kinematic errors experienced during this adaptation process. The addition of an assist-as-needed robot results in two adaptive interacting subsystems: the robot and the human. Each subsystem is composed of an optimizing controller and a next-step estimator. The assist-as-needed robotic controller mimics the structure of the adapting human, and it is configured such that it relaxes its assistance faster than the human learns. This allows it to challenge the human yet limit performance errors in an assist-as-needed manner.
Figure 1 Conceptual overview of human-robot cooperative, assist-as-needed gait training. The goal of this type of training is to allow the human to learn to compensate for the gait impairment, but to reduce the kinematic errors experienced during this adaptation process. The addition of an assist-as-needed robot results in two adaptive interacting subsystems: the robot and the human. Each subsystem is composed of an optimizing controller and a next-step estimator. The assist-as-needed robotic controller mimics the structure of the adapting human, and it is configured such that it relaxes its assistance faster than the human learns. This allows it to challenge the human yet limit performance errors in an assist-as-needed manner.CC BY · PubMed Central
Figure 2 Error weighting function. We designed a weighting function, β(ei), which scales the feedback error contribution to the robotic assistance force. If the step height error falls within the 2δ (6σ) window then the robot assistance decays as a function of equation 18. The transition zone defined by W = 1/2 σ in equation 18, allowed a smooth transition from 0 to 1 using a weighted hyperbolic tangent function.
Figure 2 Error weighting function. We designed a weighting function, β(ei), which scales the feedback error contribution to the robotic assistance force. If the step height error falls within the 2δ (6σ) window then the robot assistance decays as a function of equation 18. The transition zone defined by W = 1/2 σ in equation 18, allowed a smooth transition from 0 to 1 using a weighted hyperbolic tangent function.CC BY · PubMed Central
Figure 3 Experimental Apparatus for Creating a Virtual Impairment and for Assisting in Motor Learning. Picture (left) and diagram (right) of experimental setup. The robot makes use of a linear motor with two forcer coils and a V-shaped linkage to accommodate and drive motion of the robot's apex in the parasagittal plane. The apex is attached through a padded brace and revolute joint to the subject's lower shank. Subjects wore a harness attached to an overhead frame to catch them in case they fell. The robot created a virtual impairment by pushing upward on the subject's lower shank during the swing phase of gait, making the subject step abnormally high. The robot also assisted the subject in learning to compensate for the virtual impairment with small step height errors by helping counteract the impairment.
Figure 3 Experimental Apparatus for Creating a Virtual Impairment and for Assisting in Motor Learning. Picture (left) and diagram (right) of experimental setup. The robot makes use of a linear motor with two forcer coils and a V-shaped linkage to accommodate and drive motion of the robot's apex in the parasagittal plane. The apex is attached through a padded brace and revolute joint to the subject's lower shank. Subjects wore a harness attached to an overhead frame to catch them in case they fell. The robot created a virtual impairment by pushing upward on the subject's lower shank during the swing phase of gait, making the subject step abnormally high. The robot also assisted the subject in learning to compensate for the virtual impairment with small step height errors by helping counteract the impairment.CC BY · PubMed Central
Figure 4 Experimental protocol and results for a representative subject. This subject (30 year old male) participated in each of the three experimental protocols – A: Assistance-as-needed (AAN), B: AAN with Noise Insensitivity, and C: AAN with Improper Parameter Selection. For each protocol, the subject first adapted to the virtual impairment without assistance (steps 191–290). The after effect of adaptation, indicative of internal model formation, is apparent at step 291 in A, B, and C. Then, following a period in which adaptation was washed out (steps 291–390), the subject adapted again to the virtual impairment (steps 391–591), but with the form of robotic assistance associated with the particular protocol. In all three protocols, the step height errors when the virtual impairment was turned on were much smaller (step 391, A, B, C). The robot assistance decreased for the basic AAN algorithm, but not all the way to zero (E, trace labelled "A"). The robot assistance decreased to zero for the AAN with Noise Insensitivity algorithm (E, trace labelled "B"). The robot took over the task for the AAN with Improper Parameter Selection algorithm (i.e. with fR > fH) (E, trace labelled "C"). Note that the virtual impairment force (D) was unexpectedly turned on or off ("catch trials") to measure direct effects (steps 40–140) and after effects (steps 491–591).
Figure 4 Experimental protocol and results for a representative subject. This subject (30 year old male) participated in each of the three experimental protocols – A: Assistance-as-needed (AAN), B: AAN with Noise Insensitivity, and C: AAN with Improper Parameter Selection. For each protocol, the subject first adapted to the virtual impairment without assistance (steps 191–290). The after effect of adaptation, indicative of internal model formation, is apparent at step 291 in A, B, and C. Then, following a period in which adaptation was washed out (steps 291–390), the subject adapted again to the virtual impairment (steps 391–591), but with the form of robotic assistance associated with the particular protocol. In all three protocols, the step height errors when the virtual impairment was turned on were much smaller (step 391, A, B, C). The robot assistance decreased for the basic AAN algorithm, but not all the way to zero (E, trace labelled "A"). The robot assistance decreased to zero for the AAN with Noise Insensitivity algorithm (E, trace labelled "B"). The robot took over the task for the AAN with Improper Parameter Selection algorithm (i.e. with fR > fH) (E, trace labelled "C"). Note that the virtual impairment force (D) was unexpectedly turned on or off ("catch trials") to measure direct effects (steps 40–140) and after effects (steps 491–591).CC BY · PubMed Central
Figure 5 The first hundred steps with robotic assistance. The steps shown correspond to steps 391–491 in Figure 4. The values from Figure 4E are made negative to better visualize the decay of robotic assistance. A: Assistance as Needed protocol. Each line is data from one of the ten subjects. B: Assistance as Needed with Noise Insensitivity protocol. This assistance algorithm took into account variability in stepping height around the desired mean. Here we used a weighed error band and decreased the importance of errors within a range of normal fluctuation. When subjects exited this range, the robot assistance transiently increased C: Assistance-as-needed with Improper Parameter Selection protocol. We set the robot forgetting factor greater than the human forgetting factor (fR > fH). This created a situation in which the robot took over the task of compensating for the virtual impairment from the subjects, not allowing the subjects to learn a model of the virtual impairment. Four subjects participated in this last protocol.
Figure 5 The first hundred steps with robotic assistance. The steps shown correspond to steps 391–491 in Figure 4. The values from Figure 4E are made negative to better visualize the decay of robotic assistance. A: Assistance as Needed protocol. Each line is data from one of the ten subjects. B: Assistance as Needed with Noise Insensitivity protocol. This assistance algorithm took into account variability in stepping height around the desired mean. Here we used a weighed error band and decreased the importance of errors within a range of normal fluctuation. When subjects exited this range, the robot assistance transiently increased C: Assistance-as-needed with Improper Parameter Selection protocol. We set the robot forgetting factor greater than the human forgetting factor (fR > fH). This created a situation in which the robot took over the task of compensating for the virtual impairment from the subjects, not allowing the subjects to learn a model of the virtual impairment. Four subjects participated in this last protocol.CC BY · PubMed Central
Figure 6 Direct effect and after effects. Comparison of direct effect and after effect sizes as a function of robotic assistance. Asterisk (*) indicates a significant difference, p < 0.05, as measured by a one-way ANOVA between samples. The first two AAN experiments had ten subjects. The third experiment had four subjects. Standard error bars are shown for all means.
Figure 6 Direct effect and after effects. Comparison of direct effect and after effect sizes as a function of robotic assistance. Asterisk (*) indicates a significant difference, p < 0.05, as measured by a one-way ANOVA between samples. The first two AAN experiments had ten subjects. The third experiment had four subjects. Standard error bars are shown for all means.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 39 references

Research works printed but not linked: 3

  1. Trombly CA: Occupational therapy for dysfunction, 4th Edition. 4th edition. Baltimore, Williams and Wilkins; 1995.
  2. Kahn LE, Lum PS, Reinkensmeyer DJ: Selection of Robotic Therapy Algorithms for the Upper Extremity in Chronic Stroke: Insights from MIME and ARM Guide Results: ; Kaist, Daejeon, Republic of Korea. ; 2003:208-210.
  3. Reinkensmeyer DJ, Emken JL, Liu J, Bobrow JE: The Nervous System Appears to Minimize a Weighted Sum of Kinematic Error, Force, and Change in Force when Adapting to Viscous Environments during Reaching and Stepping: Oct; San Diego, CA,. ; 2004.

Cited by

See all 200

Metadata

Authors
Emken, J. L., Benitez, R., & Reinkensmeyer, D. J.
Affiliations
University of California, Irvine (US)
Corresponding author
Reinkensmeyer, David J
Venue
Journal of NeuroEngineering and Rehabilitation
Published
2007-03-28
Volume
4
Issue
1
Pages
8-8
ISSN
1743-0003 (ISSN-L)
OpenAlex
W2003775018
Cited by
199
References
39
Citations by year
2026: 7 · 2025: 12 · 2024: 8 · 2023: 10 · 2022: 8 · 2021: 11 · 2020: 12 · 2019: 9 · 2018: 11 · 2017: 10 · 2016: 8 · 2015: 10 · 2014: 13 · 2013: 14 · 2012: 15
FWCI
7.18
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Journal of NeuroEngineering and Rehabilitation (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RECERCAT (Consorci de Serveis Universitaris de Catalunya) (acceptedVersion, other-oa)
  • LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) (submittedVersion, public-domain)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
  • UPCommons institutional repository (Universitat Politècnica de Catalunya) (submittedVersion, cc-by-nc-nd)
Funders
  • Achievement Rewards for College Scientists Foundation
  • National Institutes of Health (NIH-R01)
  • National Institute of Standards and Technology
  • National Institute on Disability and Rehabilitation Research (H133E020732)
  • National Institute of Neurological Disorders and Stroke (R01NS040917)
  • National Center for Research Resources (M-01RR00827, M01RR000827)
Topics
  • Physical Sciences › Engineering › Biomedical Engineering › Prosthetics and Rehabilitation Robotics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Motor Control and Adaptation
  • Health Sciences › Medicine › Rehabilitation › Stroke Rehabilitation and Recovery
Keywords
robotic assistance, human-robot collaboration, sensory-motor transformation, motor learning, force fields, treadmill walking, forgetting factor, error weighting, movement variability, kinematic error
MeSH terms
Adult, Algorithms, Female, Humans, Learning, Male, Models, Theoretical, Nervous System Diseases, Robotics, Task Performance and Analysis, Vision Disorders, Walking

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Human-robot cooperative movement training: Learning a novel sensory motor transformation during walking with robotic assistance-as-needed. https://app.cochat.ai/papers/10.1186/1743-0003-4-8
The paper
Emken, J. L., Benitez, R., & Reinkensmeyer, D. J. (2007). Human-robot cooperative movement training: Learning a novel sensory motor transformation during walking with robotic assistance-as-needed. Journal of NeuroEngineering and Rehabilitation. https://doi.org/10.1186/1743-0003-4-8

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.