|
Listen to this article  |
Today’s sensor-driven revolution is transforming robots from rote machines into cognitive collaborators. They have become a key link in a dynamic continuum that encompasses humans, other machines, and the digital environments in which they operate.
The potential payoffs of sensor-guided, human – robot collaboration are huge. Examples range from protecting workers and increasing productivity, to driving new revenue streams through innovation and more.
The collaborative environment enabled by the automation continuum includes diverse players and reams of data, which together can present several significant challenges. Fortunately, they can be solved using the same technologies that made the automation continuum possible. Those challenges include:
Challenge #1: Proximity to Human Collaborators
Having ‘fragile’ humans work amid powerful machines is risky. To date, this vulnerability has been minimized by simply barring humans from working near active robots, often distancing man from machine with shields and guardrails, or even putting each in completely separate rooms. But in collaborative environments, such separation tactics are no longer feasible, as humans increasingly inhabit close quarters with their partnering robots (aka collaborative robots or ‘cobots’).
In addition to mechanical safeguards, cobots rely on multiple sensors and technologies like AI to make sense of the world around them and operate safely in it. At the same time, the environment in which a new robot finds itself installed or traversing through will increasingly feature multiple sensor-intensive IoT devices, with many more on the way.
Many consider IoT and robotics technologies as separate fields, so the synergies across the two disciplines go unexplored. But reimagined together, IoT and industrial robotics becomes the Internet of Robotic Things or IoRT.
Challenge #2: Dealing with Data Overload
Higher levels of machine awareness make for an industrial environment increasingly rich in sensor-derived data, but traditional computing frameworks often can be overwhelmed, negating the benefits of a robotically enhanced workforce. Pushing data into the cloud for processing is no longer practical for many applications.
The solution rests at the edge. With artificial intelligence and access to high volumes of data, edge devices, including robots, can make decisions much faster than humans. Computing increasingly needs to take place on the edge as robots are better equipped to perform more activities and make more decisions autonomously. Productivity is increased at the edge.
A ‘self-aware’ robot driven by data gathered and processed at the edge can detect the likelihood of its own imminent breakdown.
For example, a ‘self-aware’ robot driven by data gathered and processed at the edge can detect the likelihood of its own imminent breakdown. Communicating with others on the assembly line, the at-risk machine can shut itself down while other robots adapt their workflow in real-time to make up for the missing ‘worker.’ The production line slows, but does not stop. A human collaborator can make the needed fix and the system returns to full speed.
There is an edge-driven paradox.. as robots become ever more sophisticated, capable, and dexterous, the effort required to train them declines in many cases. Leading manufacturers understand that shortening the end-user learning curve is a fundamental means of boosting the appeal of industrial robots.
Challenge #3: End-to-End Cybersecurity
As robots become more mobile, collaborative, edge-resident and connected, the data-rich ecosystem can become a target for hackers. Companies may become vulnerable to malware, cyber-ransom, production delays, and business disruption. In addition, cyberattacks targeting powerful robotic systems also pose serious, physical safety concerns.
The solution? A comprehensive, end-to-end approach to cybersecurity. System integrators need to understand the machines they are installing, and the overall environment in which they operate, with an eye toward identifying potential access points and hardening vulnerable targets. The robot operator’s IT team must be engaged, actively monitoring threats and updating security measures.
Security must also extend beyond end-of-life, eliminating the possibility of a device having an afterlife in malicious hands. Outdated edge devices occasionally show up on eBay, where hackers can buy them on the cheap, then reverse engineer them. So, it is vital to decommission devices using tamper-proofing measures or by wiping sensitive software – making it impossible to reverse engineer.
Challenge #4: Cost
Advanced technologies and new business models are driving economies of scale in robotics, which is good news considering that 53% of prospects for industrial robots see cost as their number one challenge. With the rise of Robots-as-a-Service (RaaS), more manufacturers are becoming service providers, allowing customers to scale the number of running units to accommodate demand.
AI and machine learning algorithms have become more efficient, making it easier to program robots, devise innovative use cases, and reduce energy requirements.
Emerging business models like RaaS and leasing help lower costs, removing barriers for customers to automate with robots. Robot developers and integrators can also make it easier for potential end-users to envision compelling use-case scenarios before committing to a final investment.
Computing, data communication, and storage advances deliver more for less. AI and machine learning algorithms have become more efficient, making it easier to program robots, devise innovative use cases, and reduce energy requirements. While some believe that Moore’s law no longer applies to the quantity of transistors, it continues to hold true in terms of the cost of computing, as more and more capabilities become available at lower processing prices.
Conclusion
Unprecedented advances in sensor technology, computing power, and edge processing can provide robots with robust AI capabilities, but this is predicated on secure, but flexible connectivity and interoperability among all ecosystem participants. Robots must be able to connect readily to other robots, and also with a full range of IoT, edge, cloud, and analytical tools and other devices.
To date, the robotics and IoT communities have been driven by varying, yet highly related objectives. IoT focuses on services for pervasive sensing, monitoring, and tracking, while the robotics community focuses on production action, interaction, and autonomous behavior. Fusing both fields leads to better robotics task execution.The robots have more data for analysis and AI enabled decision-making. In this way, edge computing opens the door for even closer collaboration between man and machine.
About the Author
Michel Chabroux, Senior Director, Product Management, Wind River
Michel Chabroux is responsible for the Product Management team driving technology and business strategies for Wind River’s runtime environments, including the VxWorks and Wind River Linux families of products. He has more than 20 years of industry experience including roles in technical sales, support, training and product management. Prior to joining Wind River, he was a consultant in Business Management and Information Systems working with a variety of clients. He holds a Master’s degree in Computer Science Applied to Business Administration from Universite de Lorraine.
Related Content
"machine" - Google News
April 02, 2021 at 05:41AM
https://ift.tt/3wid1td
The Future of Robotics: Collaboration Between Man and Machine - Robotics Business Review
"machine" - Google News
https://ift.tt/2VUJ7uS
https://ift.tt/2SvsFPt
Bagikan Berita Ini
0 Response to "The Future of Robotics: Collaboration Between Man and Machine - Robotics Business Review"
Post a Comment