Machine Learning-Based Robotics in Unstructured Environments

Frost & Sullivan, December 2006, Pages: 79

This Frost & Sullivan research titled Machine Learning-Based Robotics in Unstructured Environments provides an in-depth analysis of the technical developments surrounding learning and teach methodologies adoption in service and networked robotics. It looks at its impact on intelligent robots through key drivers, industry challenges, and patent analysis.

Technology Sectors

Expert Frost & Sullivan analysts thoroughly examine the following technology sectors in this research:

- Industrial automation and process control
- Industrial robotics
- HMI
- Sensors

Technologies

The following technologies are covered in this research:

- Sensor technology
- Speech and face recognition
- Radio frequency identification (RFID)

1. Executive Summary
- 1. Scope & Methodology
-- 1. Scope of the Research Service
-- 2. Methodology
- 2. Key Research Findings
-- 1. Technology Snapshot
-- 2. Noteworthy Emerging Technologies and Applications
-- 3. Analyst Insights

2. Technology; Applications--Viewpoint and Roadmap
- 1. Technology Primer
-- 1. The Essential Components of Machine Learning
-- 2. Common Learning Problems
-- 3. Key Types of Machine Learning
-- 4. What Robots Should Learn ?
-- 5. Robot Learning--A Difficult Machine Learning Problem
-- 6. Sensor Technology for Machine Learning-Based Robots
-- 7. Significance of Feature Identification for Navigation in Unstructured Environments
-- 8. The Different Sensors To Be Utilized
-- 9. Hardware Issues in the Design of Machine Learning-Based Robots
-- 10. Paradigms in Robot Learning
-- 11. Technology Roadmap of Machine Learning-Based Robotics
- 2. Applications Viewpoint
-- 1. Service Robotics
-- 2. Networked Robotics
- 3. Applications Roadmap
-- 1. Service Robotics
-- 2. Networked Robotics

3. Technology Adoption Factor Analysis
- 1. Service Robotics
-- 1. Technology Drivers
-- 2. Technology Challenges
- 2. Networked Robotics
-- 1. Technology Drivers
-- 2. Technology Challenges

4. Assessment of Global Research and Innovations
- 1. Research Work at Universities
-- 1. Standford AI Robot (STAIR) Project
-- 2. Cognitive Agent that Learns and Observes (CALO) Project
-- 3. Robotic Soccer: The Brainstormers--University of Osnabruck; Germany
-- 4. Intelligent Robot Systems for Elimination Units for Marine Oil Pollution (EU-MOP) Project--Greece
-- 5. Cognitive Systems for Cognitive Assistants (CoSy)--Germany
-- 6. Swarm Robotics; Universite Libre de Bruxelles--Belgium
-- 7. Home Environment Cleaning Thoroughly Operating Robot (HECTOR) Project--Germany
-- 8. Human Observation-Based Motion Control Strategies in Intelligent Space--Japan
-- 9. Multimodal Teleoperation Interface for a Mobile Robot Based on Ubiquitous Information Access--Japan
-- 10. Toward a Touching Presence--Technische Universitat Munchen; Germany
- 2. Work at Corporate Establishments
-- 1. Learning Applied to Ground Robots (LAGR)--USA
-- 2. Autonomous Navigation Technology--Switzerland
-- 3. Corporate Contributions
-- 4. International Comparisons

5. Directory of Patents and Key Contacts
- 1. Key Patents
-- 1. List of Key Patents--2006
-- 2. List of Key Patents--2005
- 2. Contacts
-- 1. Universities
-- 2. Corporates

6. Decision Support Database
- 1. Decision Support Database Tables
-- 1. Number of Manufacturing Units--(1999-2006)
-- 2. Number of Service Organizations--(1999-2006)
-- 3. Number of Households--(1999-2006)
-- 4. Labor Force Population--(1999-2006)
-- 5. Number of Persons Employed in Manufacturing Industry--(1999-2006)
-- 6. Number of Persons Employed in Service Industry--(1999-2006)

List of Figures

Chapter 2
Supervised learning
Unsupervised learning
Reinforcement learning
Traditional decomposition of an intelligent control system
New approach to an intelligent control system
Application of networked robotics
Roadmap to personal robots
Network robotics--An approximate timeline

Chapter 3
Technology challenge roadmap--Components
Technology challenge roadmap--Advanced behaviors
Network robotics research challenges
Technical challenges faced in network robots

Technology Overview

The Growth of Learning-based Robotics is Dependent on Development in Related Hardware

The possibility of building learning systems, which can operate on realistic robots, is a challenging task since learning-based robots need clear sensing/perception capabilities and related hardware. "Although hardware is evolving and promising, software mechanisms still dominate in learning," according to the analyst of the study. "Since machines cannot be made 'ready to go' in a complex environment, they are likely to be improved by implementing on-board learning skills."

Service robotics, which includes industrial, service, and personal robots, is a rapidly emerging market. Despite this, the growth rate is restrained by lack of significant capital investment. This market demands high entry fees and investment from every industrial robot manufacturing entrant, which reduces the participant’s profit margin. Ultimately, size, shape, mobility, interaction, and safe operation are crucial for the success of service robotics that is implemented in natural environments. These features, however, depend on developments in related hardware fields.

Consumers Interest in Intelligent Systems Drives Advances in Learning-based Robotics

The need for flexible production systems that can be applied in non-structured environments and current interest in building intelligent systems have driven research on networked robotics. As the robot network begin to function in an unstructured environment, visualization, mapping, sensing, and information processing change from the structured to the unstructured environment. Hence, there is a pressing need to address dynamic topology management in networked robots.

The current trend demands that robots possess the ability to identify and plan a sequence of action, plan actions independent of their actual execution, and have the capability to modify or abandon plans. "Autonomous acquisition and execution in robotics require robots to learn to operate and undertake decisions autonomously," explains the analyst. "In addition, developments in artificial intelligence (AI) have led to the incorporation of intelligence and common sense in robots to help them work in changing and unstructured environments."

Customers who bought this item also bought

• Service Robotics - Global Strategic Business Report
• Industrial Robotics Market - Global Forecast & Analysis By Applications, Functions, Product & Geography (2012 - 2017)
• Industrial Robotics - Global Strategic Business Report
• Global Industrial Robotics Market Forecast & Opportunities, 2018
• Educational and Entertainment Robot Market Strategy, Market Shares, and Market Forecasts, 2008-2014
• Computer Assisted Surgical (CAS) Systems - Global Strategic Business Report
• 2011-12 Germany Small Business Market Opportunity Assessment
• Global Service Robotics Market (2010 - 1015)
• 2011-12 Germany Business Market Opportunity Assessment
• Snake Robots Market Shares, Strategies, and Forecasts, Worldwide, 2012 to 2018