introduction
The research and development of robots is one of the hot topics in academia and industry today. The number of robots released has increased significantly compared to the past. The robot market size is expected to reach 470 billion US dollars in 2010. In the past, people could only see robots in laboratories or high-tech competitions, but now they have entered industrial applications and even the lives of ordinary people[1]. At present, the fields covered by robot applications include driverless vehicles, emergency rescue, personal and service robots, medical robots, agriculture and mining, education and intellectual development, etc., and with the development of science and technology, the application fields of robots are constantly expanding.
However, the rapid development of robots has also brought enormous challenges to researchers and developers in this field. Dr. David Barrett, a renowned scientist in the field of robotics (former Vice President of iRobot, Director of Disney's DreamWorks, and currently the SCOPE Program Director at Olin College), summarized three major challenges facing robot development at the 2009 NIWeek Graphical Systems Design Conference: creating smaller, lighter, and more powerful batteries; creating smaller, lighter, and more powerful actuators; and the need for an industrial-grade software development platform with good compatibility. These challenges have also resonated with other industry leaders: Dean Kamen, inventor of the Segway and founder of FIRST and DEKA, expressed concerns about traditional embedded methods in robot development; Ellen Purdy, Director of Ground Robotics and Autonomous Systems at the U.S. Department of Defense, focused on the need for further standardization and durability (battery capability) of autonomous systems; and Dr. Hee Chang Moon of the Pohang Intelligent Robotics Institute in South Korea emphasized the need for integrated embedded systems and intuitive development software. Recognizing these challenges, industry professionals around the world are striving to find solutions: Dean Kamen's team chose NI LabVIEW software and the NI CompactRIO embedded hardware platform for the development of the robot's autonomous control system in the FIRST project; TORC Technologies switched from traditional programming tools to the NI LabVIEW software platform to design its complex autonomous systems, and its CEO, Michael Fleming, spoke specifically at the NIWeek conference about how TORC benefited from powerful, advanced software, which helped simplify complex problems in system design.
Next, we will analyze these challenges and explore corresponding solutions.
Figure 1. The FIRST Robotics Competition Platform uses NI LabVIEW software and embedded hardware.
ComctRIO, with the help of these tools, enables even high school students to solve complex problems in robot design.
The challenge is to create smaller, lighter, and more powerful batteries.
Many robots need to operate independently for extended periods in harsh or outdoor environments, requiring a power source capable of providing continuous, independent power. Currently, numerous companies and research institutions worldwide are dedicated to this area of research. Many manufacturers have chosen the NI platform for various stages of development, including design simulation, production optimization, and verification testing, driving the research, development, production, and application of their innovative products. These innovative products include novel fuel cells, solar cells, and high-efficiency power generation and storage devices based on small, mobile platforms.
Robotic systems capable of long-term independent operation require not only more efficient new batteries but also optimized energy usage. This includes effectively monitoring power consumption and selecting different power-saving modes through enhanced power management features to truly leverage the performance advantages of new batteries. NI plans to add built-in power self-diagnostics and monitoring capabilities to its embedded hardware platforms CompactRIO and SingleBoard RIO, allowing robot developers to directly access battery status via corresponding APIs and design more optimized standby and power-saving modes. NI hopes that these efforts will ultimately eliminate the constraints of power issues on robot development.
Challenge 2. Smaller, lighter, and more powerful actuators
Due to considerations of robot size and weight, it is generally desirable for actuators to be lighter and smaller; however, different motion requirements also necessitate powerful actuators. As mobility needs evolve from traditional four-wheeled systems to novel biomimetic mechanics, the demand for precision actuators has increased significantly. Numerous specialized manufacturers worldwide have been engaged in research and development in this area, continuously launching high-performance products.
For robotics applications, the ultimate goal of creating smaller, lighter, and more powerful actuators is to incorporate them into robot designs. Furthermore, the powerful capabilities of actuators often require complex programming to be fully realized. Therefore, as actuators become increasingly sophisticated and complex, achieving high-precision and high-complexity control through programming has become a key issue in robot system development. Many developers in the robotics field lack expertise in motion control; in such cases, the interactive control development of actuators becomes particularly important. Leveraging over 25 years of experience connecting actuators, NI continuously reduces the complexity of control development while enhancing interactivity and flexibility by providing hardware interface modules across various platforms and interactive software tools. NI also collaborates with leading manufacturers in the industry, such as Maxon Moter, to simplify the programming development of their products.
Furthermore, the coordination between motors is crucial when constructing actuators for complex motion systems. For example, steering in a four-wheel drive system or maneuvering in a robotic arm requires the coordinated operation of multiple motors. Using the Steering and Kematics function groups provided in the NI LabVIEW Robotics module, the required rotational speed for each motor in these applications can be directly calculated, thus helping robot designers shorten the distance between high-performance motors and innovative robot applications.
Figure 2 shows that the NI embedded hardware platform CompactRIO provides a convenient motor interface and allows for interactive software development via LabVIEW.
Figure 3 shows that the FOC algorithm implemented based on LabVIEW FPGA can further improve the performance and efficiency of existing motor control systems.
Challenge 3. The need for industrial-grade software development platforms
Beyond hardware, researchers in the robotics field also require powerful software to design their autonomous systems. This is often overlooked, with many believing that "designing anything from scratch is relatively simple," but this is not the case in today's rapidly evolving robotics era. In fact, as some technological fields converge and de facto standards emerge, innovation no longer means starting from scratch; more innovative applications are achieved by organizing modular algorithms and functions. The robotics field is no exception, requiring standardized, industry-grade software to provide readily available algorithms and tools, thereby helping researchers innovate more quickly.
1. The software must be intuitive.
Many robotics developers have backgrounds in mechanical or electrical engineering, but lack the time and energy to learn complex computer software technologies. Therefore, they require a software development tool that integrates commonly used functions and algorithms, while also possessing good interactivity and ease of use. Furthermore, the end users of autonomous systems are often rescue personnel (such as firefighters), soldiers, or the elderly, who may not necessarily have computer knowledge; thus, the user interface must also be sufficiently intuitive.
2. The component must have good integration with the hardware.
Every autonomous system must sense or perceive its surroundings and react accordingly. Sensing systems require external sensors (such as laser rangefinders and sonar sensors), while reacting relies on various types of actuators. Therefore, software tools must be able to directly drive various sensors and actuators. Beyond sensing and control, decision-making and planning often need to be implemented on real hardware, meaning the software must have strong interoperability with real-time systems, real-time embedded hardware, and even FPGAs. Many current robotics software development kits lack this capability; they can simulate and run on development machines, but lack the ability to execute on real-time hardware targets.
3. The software must have good interactivity.
Designing a robot is no simple task; it requires repeated modifications and prototyping. Therefore, researchers in the robotics field need a software package that can be easily debugged, enabling both intuitive simulation and rapid implementation on real-time hardware systems to test algorithms and real I/O. After each hardware experiment, researchers may return to the development environment for further code optimization or adjustments. All of this necessitates software development tools with excellent interactivity.
Addressing the three requirements mentioned above, NI LabVIEW software is an ideal choice for robot development. LabVIEW's graphical programming interface makes it inherently intuitive. The latest LabVIEW Robotics module also integrates commonly used search, obstacle avoidance, path planning, and dynamics algorithms, maximizing the assistance for robot developers in quickly implementing applications. Furthermore, seamless hardware integration is a key feature of LabVIEW. Various sensors and actuators can be connected directly through the LabVIEW software. LabVIEW code that completes algorithm simulation can be directly downloaded to the NI embedded real-time hardware platform CompactRIO or SingleBoard RIO for execution, achieving true seamless hardware integration. The LabVIEW Robotics module also provides ready-made drivers for various commonly used robot sensors on the real-time hardware platform, simplifying robot development. Finally, LabVIEW offers an interactive debugging environment and rich interface design tools, allowing various algorithms to be implemented directly on real-time embedded hardware and easily modified and verified repeatedly, meeting the debugging needs of robot researchers. Dr. Barrett, the SCOPE project manager, stated that LabVIEW is exactly the software the industry needs: "I have spent 15 years trying to find the best robot programming language, and NI LabVIEW truly deserves the honor. It can handle various sensors, multiple actuators, complex dynamic control algorithms, and can easily verify and debug in a real-time environment to complete system implementation."
Beyond its inherent flexibility and openness, NI LabVIEW offers several advantages. For instance, the LabVIEW Robotics module provides mature pathfinding algorithms that users can reuse. Users can also add their innovations or latest research findings to these basic algorithms to create novel search methods. Furthermore, users can integrate these algorithms with other LabVIEW-provided algorithms or third-party technologies to develop more innovative applications, such as robotic rescue or medical assistance. LabVIEW is open software that supports hardware platforms from many vendors and custom devices, and it supports calling code or models generated by various third-party development tools, expanding its application scope to all areas of robotics.
in conclusion
With the rapid development of the robotics field, developers face numerous challenges. Research institutions and related manufacturers worldwide are researching ways to address these challenges. In fact, tools already exist to help robotics researchers overcome these challenges, such as NI LabVIEW software and embedded hardware platforms. We firmly believe that specialized knowledge should not be a bottleneck in creating innovative robotic applications, and that every engineer should be able to confidently develop high-end robots with the help of powerful design tools. As technology innovates and advances, many problems will be solved, while new problems will arise. However, with flexible and open development tools, these problems can be continuously solved, enabling the robotics field to reach new heights.
