The Intricacies of Robotics Design: A Comprehensive Exploration
Introduction
In the realm of technology, robotics has captured the imagination of students, entrepreneurs, and the entertainment industry alike. Advanced Micro Devices (AMD), through its interactions with diverse roboticists worldwide, has identified three fundamental principles that underpin successful robotics design: the critical role of timing, the synergistic relationship between components, and the advantages of distributed systems over centralized architectures.
Timing: The Essence of Automation
Just as location holds paramount importance in real estate, timing plays an equally crucial role in automation. Unlike conventional video calls, which rely on “best-effort” internet connections, robotic automation demands a higher level of synchronization. This synchronization stems from determinism in communications and action, ensuring that the robot’s brain seamlessly directs its arm movements or facilitates collaboration between multiple robots in tasks such as rotating and painting sheet metal.
In the world of automation, time is synonymous with money. Determinism and latency are the two halves of the timing equation. Determinism ensures that events occur as intended, without fail, while latency minimizes the time taken for a reaction to occur in response to sensor input.
Rapid processing of sensor inputs, data analysis, and mechanical action is of utmost importance in industrial settings. The temporal value of data can quickly slip away if not acted upon promptly. Robert Huber, co-founder of Radmantis, exemplifies this concept. Radmantis, an aquaculture startup, utilizes vision AI and underwater mechanical gates to grade and sort salmon based on weight, health, and other factors.
“Salmon are highly susceptible to stress, which impacts the quality of the end product,” explains Huber. “By harnessing real-time vision AI and controls, we ensure that the fish are handled stress-free, fulfilling restaurant orders with the highest quality.”
The Whole is Greater than the Sum of Its Parts
Many roboticists trace their origins to FIRST (For Inspiration and Recognition of Science & Technology), a non-profit organization that offers hands-on STEM programs to over half a million students in more than 90 countries. FIRST instills in students the concept of the whole being greater than the sum of its parts, a lesson learned through practical experience.
Alex Van Aacken, a former FIRST Robotics Competition participant and current mechanical engineering student at Ohio State University, recalls how his team maximized productivity by dividing into four specialized groups: design, prototyping and manufacturing, assembly, and testing. The robot itself was also modularized, comprising the base chassis, robotic arm, and connecting frame.
Van Aacken emphasizes the importance of development software that supports this modularized approach. The complexity of robots necessitates their division into subsystems, with multiple teams working independently yet collaboratively to assemble the final product. Just as an orchestra conductor harmonizes the efforts of individual musicians, the integration step is crucial in achieving the intended robotic outcome.
Distribution Triumphs over Centralization
In certain robots, subsystems possess a degree of autonomy by offloading processing tasks from the main controller. This distribution of processing offers both simplicity and performance advantages. For instance, a robotic arm subsystem may be tasked with completing a specific task without being explicitly instructed on how to do so.
The freedom and capability of the arm to execute the task more efficiently, leveraging information not readily available to the main controller, exemplify the benefits of decentralized controls. The popular international TV show, BattleBots, provides a compelling illustration of this concept.
Parker Holloway and Greg Needel, members of Team Switchback featured in recent seasons of the show, elaborate on the advantages of their robot’s articulating arm as an offloaded control subsystem. This subsystem performs rapid and independent calculations, such as matrix multiplications, to optimize arm movements.
“The matrix multiplication allows us to do motion profiling, ensuring that we always exert the maximum torque safely for the arm’s trajectory,” explains Holloway. Needel adds, “The main controller simply specifies the desired arm position, and the subsystem takes care of moving the arm as quickly as possible, given its current position and condition.”
In contrast, centralized systems may require fewer electronic components, resulting in lower initial costs. However, this approach often compromises quality due to extensive wire harnessing, introducing a single point of failure that can only be resolved by replacing an expensive centralized module or troubleshooting a labyrinth of wiring connections.
Modularity offers an additional advantage: it enables frequent and asynchronous technology refreshes of individual subsystems, avoiding the need for less frequent and more complex complete top-to-bottom redesigns in centralized systems.
Lessons from Robotics: Beyond Technology
The principles governing robotics development often resonate with everyday life. The growing interest in robotics among the youth reflects the importance of timeliness, dependability, teamwork, and the ability to break down problems into manageable chunks and assemble them into a cohesive solution.
Robotics also teaches young innovators the value of diversification and empowering each component of a system to handle its tasks autonomously. This philosophy can translate into various aspects of society, regardless of whether individuals continue their pursuit of robotics development.
About the Author
Chetan Khona, the senior director of industrial, vision, healthcare, and sciences at Advanced Micro Devices Inc. (AMD), has dedicated over 30 years to the embedded devices industry. His passion for the creative potential of embedded devices to solve design challenges in a differentiated and efficient manner stems from his time as a student at the Georgia Institute of Technology.
AMD proudly sponsors the MassRobotics Form and Function Challenge, whose winners will be announced at the Robotics Summit & Expo in May in Boston.