Design of a sensor network for the quantitative analysis of sport climbing

Abstract

We describe the design of a modular sensorized climbing wall for motion analysis in a naturalistic environment. The wall is equipped with force sensors to measure interaction forces between the athlete and the wall, which can be used by experienced instructors, athletes, or therapists, to gain insights into the quality of motion. A specifically designed triaxial load cell is integrated into each hold placement, invisible to the climber, and compatible with standard climbing holds. Data collected through the sensors is sent to an app running on a portable device. The wall can be adapted to different uses. To validate our design, we recorded a repeated climbing activity of eleven climbers with varying degrees of expertise. Analysis of the interaction forces during the exercise demonstrates that the sensor network design can provide valuable information to track and analyze exercise performance changes over time. Here we report the design process as well as the validation and testing of the sensorized climbing wall.

Keywords: instrumented climbing hold; motion analysis; performance; sensors; sport climbing.

Figure 1

Prototype of the sensorized wall. The wall is composed of three modules side by side; in this case, only the central module is equipped with sensors.

Figure 2

Sensor components, on the left, and a picture of a mounted sensor, taken from behind the wall. In the drawing, the sensing element is represented in gray and the strain gauges in black, the green and blue components are spacers, while the yellow component is a wooden disk. The green reference frame refers to the axes used in the sensor characterization, and the red point on the top left of the figure is the reference with respect to which testing offsets are measured, as described in the Experimental Design section. In the picture, the box on the left encases the acquisition board and is permanently attached to the sensing element.

Figure 3

Setup environment. (A) wall address setup; (B) wall geometry setup; (C) association of sensors to placements on the wall grid; (D) definition of hold tags and selection of holds that compose a climbing problem. In panel (C), the position coordinates of the placements originate in the lower left corner, position $(1,1)$ being the lower left placement; xPos is the horizontal position, yPos the vertical one. Once a hold is placed on the grid, the same environment displays a virtual representation of the wall and allows the user to assign a shape to the hold, chosen from a menu of pre-defined shapes. This step of the setup is not depicted in the figure, but the virtual representation that ensues is visible in panel (D).

Figure 4

User Detail environment. User data defined in this environment can be attached to a recording.

Figure 5

Data Acquisition environment. (A) User selection; (B) live recording.

Figure 6

Data Management and Recording Analysis environment. Top to bottom and left to right: record selection, visualization of the magnitude of the vector sum of all forces on holds tagged “hands” or “feet,” compared with climber body weight as inserted in the User Management environment, visualization of the magnitude of the force measured on each hold, visualization of the three components of the force measured on each hold.

Figure 7

Sensorized wall setup. The labels indicate the sequence of mandatory hand moves that was used in the test, as described in the Sensor network validation Section.

Figure 8

Force density plot on holds 2R, 2L, 3R, during the trial with Climber 4, which lasted about 9 min. Heat map proportional to the total time a given force value was measured on the hold.

Figure 9

The two top panels display the magnitude of the total force measured on holds tagged Hands and Feet, during the trial with Climber 4, which lasted about 9 min. In the top panel, the force magnitude is plotted without prior filtering, in the central panel it is low-pass filtered at $1/40$ Hz. The bottom panel shows the sum of the $F_z$ components (orthogonal to the climbing wall, positive outward) measured on all handholds (blue) and footholds (red), low-pass filtered at $1/40$ Hz.

Figure 10

Hands to total force ratio $R(t)$, defined in Eq. (1), for the 11 test climbers.