[Paper Review] Patterns of Selection of Human Movements II: Movement Limits, Mechanical Energy, and Very Slow Walking Gaits

Stuart Hagler|arXiv (Cornell University)|Jan 1, 2016

Balance, Gait, and Falls PreventionHealth Professions17 references3 citations

TL;DR

This paper develops biomechanical models to define upper and lower limits on human walking gaits using mechanical energy conservation and force constraints. It identifies a lower bound on average walking speed (~0.45 m/s) and an upper bound on step length (~1.5 m), showing that very slow walking gaits (0.45–0.6 m/s) represent a distinct gait category, with clinical speeds in neurodegenerative diseases falling within this range.

ABSTRACT

The biomechanics of the human body allow humans a range of possible ways of executing movements to attain specific goals. This range of movement is limited by a number of mechanical, biomechanical, or cognitive constraints. Shifts in these limits result in changes available possible movements from which a subject can select and can affect which movements a subject selects. Therefore by understanding the limits on the range of movement we can come to a better understanding of declines in movement performance due to disease or aging. In this project, we look at how models for the limits on the range of movement can be derived in a principled manner from a model of the movement. Using the example of normal walking gaits, we develop a lower limit on the avg. walking speed by examining the process by which the body restores mechanical energy lost during walking, and we develop an upper limit on the avg. step length by examining the forces the body can exert doing external mechanical work, in this case, pulling a cart. Making slight changes to the model for normal walking gaits, we develop a model of very slow walking gaits with avg. walking speeds below the lower limit on normal walking gaits but that also has a lower limit on the avg. walking speed. We note that the lowest avg. walking speeds observed clinically fall into the range of very slow walking gaits so defined, and argue that forms of bipedal locomotion with still lower speeds should be considered distinct from walking gaits.

Motivation & Objective

• To model the mechanical and biomechanical constraints that limit the range of human walking gaits.
• To derive lower and upper bounds on average walking speed and step length using energy and force constraints.
• To distinguish very slow walking gaits from normal walking gaits based on biomechanical limits.
• To link these limits to clinical observations in aging and neurodegenerative diseases such as Parkinson’s and Alzheimer’s.

Proposed method

• Develops a segment-based model of the human body with N segments and joints to simulate movement dynamics.
• Uses a metabolic energy model combining muscle force and velocity terms to estimate energy expenditure.
• Applies mechanical energy conservation to model energy loss and external work during walking.
• Derives a lower limit on walking speed by analyzing mechanical energy restoration during gait.
• Establishes an upper limit on step length by constraining maximum force the stance leg can exert on the torso.
• Validates models against empirical data from Atzler & Herbst (2005) on walking with external loads.

Experimental results

Research questions

• RQ1 What biomechanical constraints define the lower limit of average walking speed in normal gait?
• RQ2 What upper limit on step length emerges from maximum force generation in the stance leg?
• RQ3 How do models of mechanical energy loss and external work define a distinct category of very slow walking gaits?
• RQ4 Why do clinical walking speeds in Parkinson’s and older adults fall within a specific range below normal gait?
• RQ5 How do changes in force-generating capacity (e.g., due to aging or disease) affect walking speed and step length limits?

Key findings

• The lower limit on average walking speed for normal gait is approximately 0.45 m/s, derived from mechanical energy loss and restoration constraints.
• The upper limit on average step length is 1.5 m, derived from a maximum force capacity of 370 N in the stance leg.
• Atzler & Herbst’s subject failed to walk with F_ext = 160 N at step lengths >0.90 m, consistent with the model’s force limit.
• The model predicts that increased external load shifts the effective step length limit downward, explaining walking difficulties under load.
• Very slow walking gaits (0.45–0.6 m/s) are biomechanically distinct from normal gait and include clinically observed speeds in aging and neurodegenerative disease.
• Declines in walking speed and step length in aging or disease are consistent with reduced maximum force generation (F_max), leading to shorter steps and slower gait.