The journal Science Robotics ran a survey about the challenges in robotics. An invited panel of experts then sifted through the responses to come up with 10 biggest challenges that might have major breakthroughs in the next 5-10 years. Guang-Zhong Yang, director of the Hamlyn Centre for Robotic Surgery at Imperial College London, led the panel.
The robotics industry has many challenges, and this list admittedly isn’t exhaustive. Most of the challenges surround enabling technologies such as artificial intelligence (AI), perception, power sources, etc. Of course, ethics also made the list.
Here are each of the 10 biggest challenges facing the robotics industry, according to Science Robotics, and what is being done to solve them. Do you agree with these 10 challenges? What issues would you have included on the list? Let us know your thoughts in the comments.
1. New materials, fabrication methods
Gears, motors, and actuators are fundamental to today’s robots. But tremendous work is already being done with artificial muscles, soft robotics, and assembly strategies that will help develop the next generation of autonomous robots that are multifunctional and power-efficient.
2. Creating bio-inspired robots
Robots inspired by nature are becoming more common in robotics labs. The main idea is to create robots that perform more like the efficient systems found in nature. But the study says the major challenges involved with this area remain largely unchanged for the last 30 years – a battery to match metabolic conversion, muscle-like actuators, self-healing material, autonomy in any environment, human-like perception, and computation and reasoning.
Materials that couple sensing, actuation, computation, and communication must be developed and shared before this segment takes off. These advances could lead to robots with features such as body support, weight reduction, impact protection, morphological computation, and mobility.
3. Better Power Sources
Robots, typically, are very energy-inefficient. Especially for drones and mobile robots, improving the battery life is a major issue. Thankfully, increased adoption of these systems is leading to new battery technologies that are affordable, safe and offer longer cycle lives.
Work is certainly being done to make the components of a robot more power efficient. But the study mentions robots that need to operate wirelessly in unstructured environments will eventually extract energy from light, vibrations, and mechanical movement.
Research is also being done to improve battery technology beyond the nickel-metal hydride and lithium ion options currently available.
4. Communication in robot swarms
Robot swarms are tricky because they need to sense not only the environment, but also each robot in the swarm. They need to communicate with the other robots, too, while acting independently.
Perception-action loops are fundamental to creating autonomous robots that function in unstructured environments. Robot swarms require their communication ability to be embedded in this feedback loop. Thus, perception-action-communication loops are key to designing robot swarms. There are currently no systematic approaches for doing this across large groups.
The study says, however, that falling prices and increasing performances of sensors, processors, storage devices and communications hardware will lead to significant advances of robot swarms in the next 5-10 years.
5. Navigating unmapped Environments
There has been significant progress made when it comes to robots perceiving and navigating their environments. Just look at self-driving cars, for example. Mapping and navigation techniques will continue to evolve, but robots will be required to operate in environments that are unmapped and poorly understood.
Some of the challenges here include:
- How to learn, forget, and associate memories of scenes both qualitatively and semantically
- How to surpass purely geometric maps to have semantic understanding of the scene
- How to reason about new concepts and their semantic representations and discover new objects or classes in the environment through learning and active interactions
“For navigation, the grand challenge is to handle failures and being able to adapt, learn, and recover. For exploration, it is developing the innate abilities to make and recognize new discoveries,” the study says. “From a system perspective, this requires the physical robustness to withstand harsh, changeable environments, rough handling, and complex manipulation. The robots need to have significant levels of autonomy leading to complex self-monitoring, self-reconfiguration, and repair such that there is no single point of complete failure but rather graceful system degradation. When possible, solutions need to involve control of multiple heterogeneous robots; adaptively coordinate, interface, and use multiple assets; and share information from multiple data sources of variable reliability and accuracy.
6. AI that can reason
The study calls AI the “underpinning technology for robotics,” but acknowledges that “we still have a long way to go to replicate and exceed all the facets of intelligence that we see in humans.” The key is to the combine advanced pattern recognition and model-based reasoning to develop AI that can reason and has common sense.
AI that can learn complex tasks on its own with a minimum of initial training data is also critical. The study does mention that DeepMind’s AlphaGo Zero system is a good example of this, but it says “we do not yet have systems that can do this easily across heterogeneous tasks and domains.”
7. Brain-computer interfaces
Brain-computer interfaces (BCIs) enable some device and machines to be controlled by your mind. BCIs could be quite useful in augmenting human abilities in the future, but developing the technology for wider adoption is the challenge.
The equipment for sensing brain signals is expensive and cumbersome, and the data processing can be tricky. There’s also a long period of training, calibration and learning are required.
But this is certainly an exciting area to watch. Johnny Matheny, who lost his arm to cancer in 2005, is the first person to live with an advanced mind-controlled robotic arm. In December 2017, researchers from Johns Hopkins Applied Physics Lab delivered the arm to Matheny at his home in Port Richey, Florida. Johns Hopkins has received more than $120 million from the US Defense Department to help pay for the arm’s development over the past 10 years.
8. Social robots for long-term engagement
Humans are, generally, adept at interpreting social behavior. Robots are not. The study says the three biggest challenges of building social robots that truly interact with humans are modeling social dynamics, learning social and moral norms, and building a robotic theory of mind
Most social robots to date have been designed for short interactions, which isn’t how human relationships work. Social robots must expand from moment-to-moment engagements to long-term relationships.
9. Medical robotics with more autonomy
From minimally invasive surgery, hospital optimization, to emergency response, prosthetics, and home assistance, medical robotics represents one of the fastest growing sectors. But the challenge is building reliable systems with greater levels of autonomy.
A long-term challenge is to enable one surgeon to supervise a set of robots that can perform routine procedure steps autonomously and only call on surgeons during critical, patient-specific steps. The study says “Perhaps the most significant challenge of automating any clinical task is to be able to anticipate, detect, and respond to all possible failure modes. Medical device regulation of autonomous robots will likely need to develop in a manner that balances the requirements for provably safe algorithms with compliance costs.”
Stop me if you’ve heard concerns about robot ethics before. All kidding aside, it’s a major challenge the robotics industry is well aware of. But the study breaks the ethical problems into the following five topics:
- Sensitive tasks that should require human supervision could be delegated entirely to robots
- Humans will no longer take responsibility for failures
- Unemployment and de-skilling of the workforce
- AI could erode human freedom
- Using AI in unethical ways