Technology behind Sony's Qrio: Walking Stably

The content of this page is taken from the official Sony Qrio website at http://www.sony.net/SonyInfo/QRIO/. Unfortunately the page closed at the end of March 2007 - so we saved some of the info before the site was taken down.

Part 1:
Walking Stably

Part 2:
Surviving a Fall Unscathed

Part 3:
Remembering People & Talking

Part 4:
Seeing and Acting

Part 5:
Living with People

Walking Independently on Two Feet / Adapting to Walking Surface

QRIO doesn't just walk on it's own two feet. It dances fluidly with its entire body, and it expresses a full range of emotion. Behind this impressive ability is the Intelligent Servo Actuator (ISA), a drive system for the robot's joints and limbs that involves motors, gears, a computer, a multitude of sensors, aswell as the technology that governs it all. Responding to the condition of the walking surface, it directs the robot's joints in real time, harmonizing their movement so that the entire body moves safely and sure-footedly.

Intelligent Servo Actuator (ISA)

QRIO can walk on two feet and dance dynamically. To make its arms and legs strong, and yet able to move fluidly, it was necessary to develop an entirely new joint actuator. The realization of this Intelligent Servo Actuator (ISA) made it possible to build a robot with compact body design that could move its body smoothly and dynamically. [ ISA - Actuator ] In order for QRIO to walk and dance so skillfully, an actuator was needed with the ability to produce varying levels of torque at varying rpm speeds, respond with quickness and agility, not be affected by outside forces -- and do all of this efficiently. We found a way to make a smaller actuator with broadly improved function and precision. In addition, QRIO's gears are precise, quiet, and highly dependable.		[ ISA - Sensors ] QRIO's ISA system detects the current position of its joints and, based on the posture of QRIO's body, drives the joints to their calculated target position. In addition, it determines the maximum amount of torque applied to the actuators and calculates the corresponding rise in temperature to allow QRIO to move about safely. [ ISA - Controller ] QRIO's ISA has a built-in governing computer, through which it responds quickly to requests for control commands by the bipedal robot's actuators. We used the latest microchip and software technology to make the system as small as possible.

Independant, Stable Bipedal Walking

QRIO moves with "dynamic walking". "Static walking" means the robot keeps its center of gravity within the zone of stability -- when the robot is standing on one foot, its center of gravity falls within the sole of that foot, and when it is standing on two feet it falls within a multi-sided shape created by those two feet -- causing it to walk relatively slowly. In "dynamic walking", on the other hand, the center of gravity is not limited to the zone of stability -- in fact it often moves outside of it as the robot walks. People move using "dynamic walking". QRIO's underlying movement control method is based on the Zero Moment Point (ZMP)..		..the point where the combined forces of gravity and inertia working on the robot intersect with the ground. The control system directs QRIO's movement so that its ZMP is always within its zone of stability. It also senses the pressure on the soles of the feet and the attitude of the robot's body to determine the condition of the walking surface, and adapt to changes in the surrounding environment. Using a holistic movement control system that controls the movement of the robot's body based on this data, QRIO is able to walk and move dynamically and stably.

Responding to Changes in Walking Surface

How is QRIO able to walk on an uneven surface, or one whose slope suddenly changes? It is equipped with technology that uses a wide range of sensors to detect changes in the walking surface and respond accordingly. QRIO determines the condition of the walking surface using four pressure sensors in the sole of each foot to gather data on the amount of force being received from the walking surface.		For example, when the foot touches an uneven surface, the sensors detect the slope of the ground; the foot is placed down in accord with the slope; and the attitude of the body is adjusted using data from position sensors so that the robot maintains a stable position as it moves. It can adjust to disparities in elevation up to 1 cm, and slopes up to 10 degrees.