Menu
Home
Log in / Register
 
Home arrow Engineering arrow Designing for Emerging Technologies
Source

Acknowledgement of people

A human workmate’s confidence in a robot will increase if the robot acknowledges her presence. It requires addressing two separate design challenges: detection and then communication. Many different sensors can detect the presence of a person, such as cameras, lasers, infrared, sonar, capacitive, time-of-flight, and pressure sensitive resistors. The choice of sensors to employ depends on the application, environment, and cost. Decisions on what combination of sensors to use are driven by cost, precision, and environment. Table 6-1 provides a high-level description of the trade-offs.

Table 6-1. Sensors and use factors

Sensor

Precision

Challenge

Cost

Lasers (LIDAR)[a]

Very high

Limited reflections

Very high

3D Kinect

High

Noise interference

High

3D via stereo 2D

High

Combine multiple views

High

2D camera

High

Position and lighting

Medium

Sonar

Medium

Noise interference

Low

Infrared

Medium

False detection

Low

Weight pressure mat

Low

Stationary

Low

[a] Light Detection and Ranging, http://www.lidar.com/

The Multisense-S7 camera shown in Figure 6-11 is an example of a 3D sensor that combines multiple 2D views and color processing, but each unit costs more than a thousand dollars.

Multisense-S7, 3D sensor— (© 2013 Carnegie Robotics LLC)

Figure 6-11. Multisense-S7, 3D sensor[] (© 2013 Carnegie Robotics LLC)

A robot’s responses to human presence are dictated by its physical platform, but they can include sounds, speech, lights, gaze, head nods, change in speed, and gestures. Some effective acknowledgements include slowing down or stopping movement, gazing in the direction of the person, and head nods, which are common social cues between people. Lights are more startling and typically indicate a status specific to the particular robot. Beeping sounds might not be heard in a noisy environment, but they can be appropriate for urgent situations if loud enough. If the robot speaks, it should also meet the expectation that it can understand human speech. Gestures require movements from a part of the robot that is not occupied in an actual task, which might not always be the case. Anything beyond a simple wave is subject to interpretation and thus might not be clear.

The Baxter robot communicates a range of social cues via common facial expressions (look ahead to Figure 6-13). Notice how clearly the message is conveyed by using eyes and eyebrows. Excluding a mouth does not sacrifice clarity of communication and avoids any false implication that the robot can speak. This is further evidence that robots communicate better when they have heads.[70], []

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel