Image Image Image Image Image Image Image Image Image Image

Bartlett School of Architecture, UCL

Scroll to top

Top

Hypergesture

Hypergesture
  • On February 19, 2021
  • https://jklucy.com/

A large-scale robotic installation consisting of two custom robot arms based on a design for a surgical robot which constrains the movement of the end effector to orbit around a single point in space. The two robots perform a dance duet together by performing movement sequences within a shared workspace. The spherical workspace echoes the Kinesphere, choreographer and dance theorist Rudolf Laban’s conception of the range of motion of a dancer’s body – a sphere formed by the reachable area from a standing position with limbs outstretched.

 

 

Movement is controlled by two stepper motors in the base which transmit rotation to the end effector through a system of belts, pulleys, chains and sprockets. The end effector is fitted with a  LED bulb to draw the attention of the viewer to the relationship between the two end effectors which share a spherical workspace. The base allows continuous rotation to imitate the ability of virtuoso ballet and contemporary dancers to perform multiple consecutive turns.

A figure represented as a set of 10 points.

The choice to use two point light sources on the extremities of the robots was influenced by the research of psychologist Gunnar Johansson into the perception of biological motion in point light displays [1], that is perceiving human or animal motion in the movement of a small number of points of light. This idea was most recently implemented in Random International’s Fifteen Points [2] through the use of robotics.

Hypergesture takes this idea further. If we reduce this down to two points of light and robots with only two degrees of freedom, can we still achieve movement that can be related to a human figure?

The relationship between a figure and a robot has the possibility to be more dynamic because a single point of light can be associated with different parts of the dancer’s body at different times during a performance.

In a series of simulations in Three.JS, different behaviours are explored to capture aspects of a dancer’s movement. Using motion capture data from Alexander Whitley’s Digital Body Project, the robots can imitate the angle between two joints in 3d space, the elevation of a dancer (standing, relevé, plié, floorwork) and the direction the dancer is facing.

 

 

You can experiment with mapping joint relationships onto the robots in the online simulation here.

Besides this, the robots create relationships between themselves by virtue of sharing a workspace. One robot towering over the other creates an impression of dominance. Fast sudden movements create impressions of aggression while gentler coordinated movements where one arm follows another creates an impression of collaboration or friendliness between the two arms.

 

 

References:

  1. Johansson, Gunnar. ‘Visual Perception of Biological Motion and a Model for Its Analysis’.Perception & Psychophysics 14, no. 2 (June 1973): 201–11. https://doi.org/10.3758/BF03212378.
  2. RANDOM INTERNATIONAL. ‘Fifteen Points II’. https://www.random-international.com/fifteen-points-ii-2019.