The editors conducted an online survey of unresolved problems in robotics and organized an expert panel of robotics experts who listed the 30 most important topics and then divided them into 10 major challenges that could The next five to 10 years have a significant impact.
Here are the top ten challenges they put forward:
1, new materials and manufacturing programs
Robotics are beginning to no longer be confined to the methods of making robots with motors, gears and sensors, and trying to integrate multiple functions in one material using technologies such as artificial muscles, soft robotics and new manufacturing methods, but these advanced Most of the manufacturing methods are still in their infancy, so integrating them is hard to come by. Multifunctional materials such as induction, motion, energy harvesting, or energy storage can make robotic designs more efficient, but combining these different attributes in one New methods need to be used on stage machines, and the new method must be a mixture of microscale and mass manufacturing techniques.Another promising research direction is material that can be adapted to the environment or self-healing over time, but this requires Do more research.
2, bionic robot and biological hybrid robot
Nature has solved many of the problems that robotics experts are trying to solve, so many are turning to biology for inspiration and are even trying to integrate life systems into their robots, but they provide themselves with the mechanical properties and biological systems that replicate muscles Energy capacity, there is still a huge bottleneck.Artificial muscle development has made great progress, but the muscle robustness, efficiency, vitality and energy density needs to be improved.Living cells can be embedded in robots can overcome the Small robots provide the power to puzzle, but they also embed biometrics such as self-healing and embedded sensing into the robot, but how to integrate these parts remains a major challenge. While more and more 'robotics' are helping people unlock big Natural secrets, but we also need to do more to understand how animals can transition between abilities like flight and swimming to build a multichannel platform.
3, energy and energy
Energy storage is a major bottleneck in mobile robotics, and the growing demand for drone, electric vehicles and renewable energy has led to advances in battery technology, but the basic challenges have not changed for many years, At the same time as the development of batteries, efforts must be made to minimize the demand for electricity from the robots and to allow them to use new sources of energy, making them two promising research areas deserving research, drawing energy from the environment and transferring energy over the air.
4, robot cluster processing tasks
Clusters of simple robots assembled into different structures to handle different tasks may be cheaper and more flexible alternatives to larger robots that perform mission-specific tasks. Smaller, cheaper and more powerful hardware Allows simple robots to perceive the surrounding environment and, in combination with artificial intelligence, simulate what can be seen in nature's flocks.However, more is needed to achieve the most effective control over groups of robots of different sizes Work - Small groups can be centrally controlled, but large groups need to be dispersed into smaller groups for further control. These robots also need to be made more robust to adapt to real-world conditions and withstand deliberate or accidental injuries. In addition, There is also a need for more research on different robots with complementary functions.
5, navigation and exploration
One of the key uses of robots is to explore places that humans can not reach, such as deep sea, space or disaster areas, which means they need to be good at exploring and navigating environments without maps, often extremely cluttered and in crisis.The main challenges include creating a Adapt, learn and recover from navigation failures and can recognize new discoveries.This requires a high degree of autonomy, allowing the robot to monitor and reset itself, while being able to utilize multiple, with different reliability and accuracy Data sources build an image of the world.
6, robotics itself
Deep learning has revolutionized the ability of machine identification patterns, but it needs to be combined with model-based reasoning to create adaptable robots that can learn from the work process. The key point here is to create artificial intelligence, which Being aware of your own limitations and learning how to learn new things creates a system that learns quickly from a limited set of data without having to learn from the millions of examples needed for deep learning It is important that our further understanding of human intelligence will be the key to solving these problems.
7, brain machine interface
The brainchild interface will enable seamless control of advanced robots, but it can also prove a faster and more natural way to communicate instructions with robots or just to help them understand the state of mind in humans. Most current brain measurements Activities are expensive and cumbersome, so it is important to be functional on simple, low-power and wireless devices, and they tend to be extended to training, calibrate, and adapt because of imprecision in reading brain activity. It remains to be seen whether they will outperform simpler techniques such as eye movements or reading muscle signals.
8, social interaction
If robots are to enter human life, they will need to learn how to deal with humans, but that's not easy because we have very few models of human behavior and we can easily underestimate the complexity of our innate humanity. Robots need to be able to perceive subtle social cues, such as facial expressions or intonations, understand the cultural and social context in which they live, and model the mental state of the person interacting with them in order to adapt to communicating with humans, whether in Short-term relationships or long-term relationships in the process.
9, medical robots
Medicine is one of the areas in which robotics can have a significant impact in the near future.Optimization of surgeons' robotic equipment is already in use, but one of the major challenges is to increase the autonomy of these systems in such high-risk environments.Automatic robots The helper needs to be able to recognize the anatomy of the human body in a variety of situations and to use situational awareness and voice commands to understand the needs of different scenarios.In surgery, automated robots can perform routine steps in the surgical procedure to make the surgeon more sick Complicated Patient Services There is also hope that micro-robots operating inside the human body will be popularized in the future, but there are still many obstacles to using such robots, including effective delivery systems, tracking and control methods, and the key is to find treatments that will improve existing methods .
10, robot ethics and safety
As the previous challenges are overcome one by one, the robot will gradually integrate into our lives, but this development will also create new ethical challenges. Most importantly, we may over-rely on the robot, which may lead to human loss of some Skills and abilities prevent us from taking control of the situation in the event of failure, and we may eventually delegate to the robot the task of requiring human supervision for moral reasons, so that in the event of failure, people will be shirking their responsibilities to the robot. It can also reduce people's right to self-determination as human behavior changes in response to some of the daily activities and constraints required for the effective work of robots and artificial intelligence.