Wearable Goniometer and Accelerometer Sensory Fusion for Knee Joint Angle Measurement in Daily Life

Abstract

Human motion analysis is crucial for a wide range of applications and disciplines. The development and validation of low cost and unobtrusive sensing systems for ambulatory motion detection is still an open issue. Inertial measurement systems and e-textile sensors are emerging as potential technologies for daily life situations. We developed and conducted a preliminary evaluation of an innovative sensing concept that combines e-textiles and tri-axial accelerometers for ambulatory human motion analysis. Our sensory fusion method is based on a Kalman filter technique and combines the outputs of textile electrogoniometers and accelerometers without making any assumptions regarding the initial accelerometer position and orientation. We used our technique to measure the flexion-extension angle of the knee in different motion tasks (monopodalic flexions and walking at different velocities). The estimation technique was benchmarked against a commercial measurement system based on inertial measurement units and performed reliably for all of the various tasks (mean and standard deviation of the root mean square error of 1:96 and 0:96, respectively). In addition, the method showed a notable improvement in angular estimation compared to the estimation derived by the textile goniometer and accelerometer considered separately. In future work, we will extend this method to more complex and multi-degree of freedom joints.

Keywords: accelerometers; data fusion; human motion analysis; joint angle measurements; knee joint; knitted piezoresistive fabrics; sensor to segment alignment; smart textiles; wearable goniometers.

Figure 1

Sensing prototype and sensor placement around the knee joint. (a) A double-layer knitted piezoresistive fabrics (KPF) goniometer and two inertial measurement units (IMUs) were applied to the knee band; (b) A geometrical scheme of the reference frames fixed with the body segments and the IMUs. The knee was simply modeled as a hinge joint, and the flexion-extension angle (θ) was defined as the angle between the two consecutive model segments (i.e., the angle between the x unit vectors of the ${\Psi }_1[x_1,y_1,z_1]$ and ${\Psi }_2[x_1,y_1,z_1]$ frames). IMU and accelerometer reference frames (${\Psi }_{a_1}[x_{a_1},y_{a_1},z_{a_1}]$ and ${\Psi }_{a_2}[x_{a_2},y_{a_2},z_{a_2}]$) are not aligned with the corresponding segment reference frame.

Figure 2

Schematic diagram of a double-layer KPF goniometer. The black stripes represent the two identical piezoresistive layers, while the gray stripe is the insulating layer. When the sensor is in the flat position, the resistance difference ($\Delta R$) between the two layers is zero. When the sensor is flexed, $\Delta R$ is proportional to the bending angle (θ), defined as the angle between the tangent planes to the sensor extremities (green dashed line in the picture).

Figure 3

The calibration procedure for the accelerometers. In the first step, using the accelerometer output only, acquired from a subject in standing position, the $x_{ai}$ axes of the accelerometer frame are aligned with the corresponding $x_i$ axis of the bone frames by computing the ${\widehat{\gamma }}_i$ and ${\widehat{\beta }}_i$ angles. In the second step, using data collected by the goniometer and the accelerometers in a dynamic acquisition, the remaining axes of the inertial frames are aligned with the corresponding axes of the frames fixed with the joint segments.

Figure 4

The goniometer/accelerometer fusion methods. The grey box represents the Kalman filter in its error state or indirect form.

Figure 5

Dynamic comparison between our estimation technique and the reference measurement during contralateral monopodalic standing tasks with different knee flexion-extension velocities. The blue line represents our estimation, while the red line is the reference measurement. (a) Slow knee flexion; (b) Fast knee flexion.

Figure 6

Dynamic comparison between our estimation technique and the reference measurement during walking at different velocities. Velocities increase from (a) to (d). The blue line represents our estimation, while the red line is the reference measurement.

Figure 7

Signal comparison between the angle reconstruction by the accelerometers (green dotted line), the goniometer (black dotted line), the hybrid system (accelerometer + goniometer, blue solid line) and the reference measurement. (a) Walking; (b) Fast flexion.