**1. Introduction**

The socket is the mechanical interface between prosthesis and residual limb. It accounts for stability, ensures control over the prosthetic device, and determines its level of comfort. Thus, stump–socket interaction strongly influences the well-being and mobility of amputees [1]. To ensure the quality of socket fit, quantitative measures are advantageous. Pistoning or relative movement between residual lower limb and prosthesis is considered to be one indicative parameter of socket fit quality [2].

The most common techniques that are used to acquire data on the relative movement between residual limb surface and prosthetic socket are: motion capture [3–6], other optical means [7,8], inductive sensors [9–12], and vacuum pressure fluctuations of elevated vacuum suspension systems [11,13]. These approaches are subject to different drawbacks. For instance, vacuum pressure fluctuations cannot be used to evaluate the relative motion at specific or problematic locations within the interface. Additionally, despite the number of different measurement approaches, studies presenting gait cycle dependent data of relative motion during ambulation are scarce [3,4,9,14].

A concept for measuring the relative motion between residual limb surface and prosthetic socket at specific locations in dynamic gait situations is presented in [15]. The proposed concept is based on optical 2D-motion sensor units, whose applicability has been tested experimentally using a functional model consisting of one sensor unit on a test rig as well as in biomechanical substitute studies [16].

This paper presents the subsequently realized low-cost measurement system capable of interfacing with an array of sensors. The implementation of the optimized measurement chain is presented in Section 2. In Section 3, the measurement system is evaluated experimentally on a test rig regarding accuracy, precision, achievable sampling frequency, and overall reliability. A pilot study using seven sensors to measure the relative motion between residual limb and prosthetic socket is reported in Section 4. Based on a discussion of the results, the paper is concluded in Section 5 where an outlook to future work is also given.
