The coming evolution of Service Robots: From Shape-Shifting Robots to Robot Inclusive Cities

There is no denying it: like it or not we are witnessing a service robotics revolution, in both professional and personal domains. From residential floor cleaning to logistics delivery missions, robotics offers enormous advantages in improving productivity, efficiency and safety in both professional and personal settings. Professional service robots are mainly used in professional settings outside the traditional home or manufacturing scenarios. While industrial robots are primarily used in the automation of manufacturing tasks, today’s rapid technological advances has resulted in the development of non-factory robotic automation of a wide range of menial, repetitive, time consuming or dangerous tasks thereby freeing human workers for engagement in more cognitive functions. According to the International Federation of Robotics (IFR), the market of the professional service robots is expected to grow at an average rate of 20 to 25% between the year 2018 and 2020, eventually hitting an estimated value of $27 billion1. On the other hand, personal service robots mainly consist of consumer based robots commonly used for automation of home related tasks some of which include autonomous window cleaners of vacuum cleaners. Although this is a comparatively smaller segment of service robots, IFR predictions indicate that personal service robots will acquire an estimated market worth of $11 billion by the year 2020.

For far too long now, much of the debate around service robots has been mainly inclined towards them stealing our jobs and we humans becoming slaves to machines. This discussion has not been really true. We have to reframe this discussion in terms of the potential of service robots. Many of these intelligent machines have the potential to mitigate the combined pressure of skyrocketing costs, aging populations in industrialized countries, and a shortage of qualified workers, as well as the need to continuously improve the quality of services and results. More market surveys on the potential of service robots have shed some promising light on the advantages such machines have as they continuously enter our social spaces and actually make lives better for humans.

With rapidly expanding application venues and continuous drive towards improving productivity, cost and safety, designing these intelligent machines are becoming increasingly challenging mainly attributed to the complex nature of the environments in which they operate and the dynamic nature of the tasks they perform. In response to these recent challenges, researchers in the field are pushing for a two-pronged approach that involves 1) designing shape-shifting robots that are able to automatically adapt their morphologies to maximize their performance and 2) designing an inclusive and friendlier environment so that robots can function more efficiently.

Shape-shifting robots are electromechanical machines with variable degrees of morphology. Beyond typical perception, actuation and control capabilities found in fixed morphology robotic platforms, shape-shifting robots are also capable of deliberately changing their physical form in order to overcome environmental constraints, undertake new tasks, or recover from damage. Such robots have tremendous potential in advancing robotics as a field in general. Their promise of high degree of versatility, robustness and modularity is set to open up a wide range of new applications for robots. To see how shape-shifting robots can boost productivity, we need look no further than Singapore University of Technology and Design (SUTD) where a research group has developed a series of seven transformative robotic solutions for cleaning and maintenance jobs supported by the National Robotics Research & Development Programme Office (NR2PO). These shape-shifting robots are proving to be more efficient in comparison to traditional fixed morphology designs in a wide array of deployment situations ranging from floor cleaning, glass façade cleaning, staircase cleaning, and inspection of drains.  

Shape-shifting robots have tremendous potential in advancing robotics as a field in general. Their promise of high degree of versatility, robustness and modularity is set to open up a wide range of new applications for robots.

One of the solutions — sTetro is a shape-shifting robot that switches between cleaning flat floor and staircases. Considering the crucial role staircases play and their permanent presence even after the advent of lifts and escalators virtually in every multi-storey building, the subject of cleaning staircases has received so little attention from roboticists. Even though there exists tens of thousands of robotic products in the market that tackle floor cleaning, they all make big claims on how well the robots avoid staircases with no viable solutions towards cleaning staircases. With staircase cleaning being more strenuous job than floor cleaning, sTetro is being developed to target this large segment within cleaning robotics industry.  sTetro is capable of autonomously navigate and clean a given space including staircases. It is equipped with sensors and software algorithms to automatically detect and adapt to different types of staircases.

Mantis is another shape-shifting robot developed at SUTD aimed at glass façade cleaning. Conventional robots deployed for façade cleaning demands human intervention to manually move them from one glass panel to another. Mantis is capable of autonomously cleaning vertical glass facades as well as move over obstacles and transiting between glass panels without any human intervention. This presents a breakthrough in surface propagation robots opening up potential new applications in construction, aircraft maintenance and infrastructure facility inspection. The prototypes are currently on trial at various parts of the SUTD campus.

Generally, the expected increase in robotics population is not inherently bad as robots are playing a critically essential role in helping with many dull, menial, repetitive, time consuming or dangerous tasks for humans. Some of the robots have proven to be highly efficient and this is a significant advantage. Recent technological advancements such as developments in machine learning, cloud, IoT and artificial intelligence have made the adoption of robots to be highly beneficial and easy. However, as we embrace the expected surge of robots, it is imperative to change and adapt our infrastructure. Consequently, policy makers, architects and town planners should prepare for this to ensure that the robots will be able to lead inclusive lives within society. This will provide a significant push to the growth of service robots in the commercial marketplace. It is widely believed that a significant majority of service robots will likely be found in cities. This will present enormous challenges and it is inevitable that the cities will need to adapt in order to accommodate the rising number of robots in the workforce.
    
To enable and empower service robots to contribute productively in the workplace, we need to rethink the designs of their workplaces. Traditionally, design of contemporary new spaces and everyday artefacts such as lighting and furniture target the majority of well able-bodied population until recently when apposite design principles were introduced in response to individuals with special needs such as children, elderly and user groups with physical and/or sensory disabilities. Since cognitive capacity, physical strength and speed, visual acuity, and auditory sensitivity are still evolving, ergonomic workplace designs that accounts for robots as stakeholders will become increasingly important. Given how robots, like human workers, are an incredibly diverse group with different needs, it will also be necessary to apply universal design principles to promote wellbeing and safety for robots of all applications. To this end, Singapore University of Technology and Design (SUTD) is developing an emerging research field “Robot Ergonomics” that bridges traditional disciplines including architecture, product design and robotics based on the premise that service robots and the everyday environments that they inhabit with humans (buildings, products, furniture, tools, etc.) are more adequately conceived when designers across disciplines work in unison. Robot Ergonomics remains a marginal research area, despite the significant growth in the adoption of service robots. In robotics, increasingly complex and autonomous systems are being developed in order to cope with everyday tasks in pre-defined physical environments. Strategies place all the responsibility on the robot by combining advanced sensors, control and actuators to achieve relatively simple capabilities such as turning a handle to open doors, when the location, shape and behavior of the handles are unpredictable and highly variant. Robot Ergonomics design significantly decreases the difficulty of such tasks by taking into consideration the characteristics of robots when designing the space without overstepping into human preferences. Bridging the decision-making in robot and spatial design carries a twofold advantage: designers and architects view robots as target stakeholders, and roboticians build upon the environmental features to make future robots highly capable at manageable costs. Through an inductive study of popular cleaning and logistics robots in the market, the team at SUTD has derived a set of design principles that can be used by practitioners to generate robot-inclusive solutions. The principles put forward offers valuable insights for ethnographic studies of such service robots and the type of effects that everyday furniture and spaces have in their performance. In the future, the way forward in designing workplaces will not be to adopt a one size fits all approach, but rather one that gives robots the flexibility to adapt the workplace to suit their own needs while prioritizing the human preferences.

As cities experience a growing robotics workforce, the need for robot friendly design is becoming ever more critical. From our hawker centers to parks, the future of urban housing and mobility may just be shaped for and by the robots that we will live with.

The ongoing work at SUTD is just an example of how workplaces can be successfully redesigned to help robots work productively. We need to expand and adapt these efforts to all other lines of work – be they food and beverage, transport, construction, or healthcare. We would need more generalized education so that it is possible for architects and roboticians to co-design robots and spaces unleashing potentially tens of thousands of robots from research labs into commercial world. Given that colleges and universities have institutional inertia in brining disciplines together, this challenge is not trivial.

Ultimately, efforts to design for a robotic workforce need to go hand in hand with other initiatives to make the workplace more robot inclusive. There needs to be a shift to more acceptance and flexible workplace arrangements involving humans and robots and a shift in attitudes about robots. The bigger challenge however, is to be cognizant with human preferences, costs and risks to develop future proof resilient robots. Robots are co-workers that make lives better for us and not there to steel our jobs, and it is time we take efforts to recognize that.

To get a better understanding of the unique capabilities of these robots to change their shape and adapt, please check out the following videos online:

hTetro – self-reconfigurable floor cleaning robot

Mantis – a highly agile glass façade cleaning robot