With recent advances in artificial intelligence and robotics technology, there is a growing interest in developing and commercializing home robots capable of handling a variety of household chores.
Tesla is building a humanoid robot, which, according to CEO Elon Musk, can be used to cook meals and help the elderly. Amazon recently acquired iRobot, a prominent robotic vacuum manufacturer, and has invested heavily in technology through the Amazon Robotics program to expand robotics technology to the consumer market. In May 2022, Dyson, a company best known for its vacuum cleaners, announced that it plans to build the UK’s largest robotics center dedicated to developing home robots that carry out daily household tasks in residential spaces.
Despite the increased interest, potential customers may have to wait a while for those bots to appear on the market. While devices such as smart thermostats and security systems are widely used in homes today, the commercial use of home robots is still in its infancy.
But building home robots is much more difficult than smart digital devices or industrial robots.
One of the main differences between digital and robotic devices is that home robots need to process things through physical contact to carry out their tasks. They have to carry dishes, move chairs, pick up soiled clothes and put them in the washing machine. These processes require the robot to be able to handle fragile, soft, and sometimes heavy objects of irregular shapes.
The latest artificial intelligence and machine learning algorithms perform well in simulated environments. But contacting things in the real world often causes them to falter. This happens because physical contact is often difficult to model and even difficult to control. While humans can easily perform these tasks, there are significant technical hurdles for home robots to reach the human ability to manipulate objects.
Robots have difficulty with two aspects of handling things: control and sensing. Many robotic controls such as those on assembly lines are equipped with a simple handle or specialized tools intended only for certain tasks such as holding and carrying a particular part. They often struggle to handle objects with irregular shapes or elastic materials, especially because they lack the effective force, or tactile feedback humans naturally have. Building a general-purpose robotic hand with flexible fingers remains a technical and costly challenge.
It is also worth noting that traditional robotic processors require a stable platform to operate accurately, but accuracy drops significantly when used with mobile platforms, especially on a variety of surfaces. Coordination of movement and manipulation of a mobile robot is an open issue in the robotics community that must be addressed before home robots with extensive capabilities can reach the market.
There is a high-end robotic kitchen on the market already, but it operates in a highly organized environment, which means that all the things it interacts with – cooking utensils, food containers, and appliances – are in the places it expects, and no annoying humans do. get in the way.
In the assembly line or warehouse, the environment and the sequence of tasks are strictly regulated. This allows engineers to pre-program the robot’s movements or use simple methods such as QR codes to locate target objects or locations. However, household items are often disorganized and placed haphazardly.
Home robots must deal with many uncertainties in their workplaces. The robot must first locate and select the target element among many others. It also often requires the removal or avoidance of other obstacles in the workspace to be able to access the item and perform the specified tasks. This requires the robot to have an excellent cognition system, effective mobility skills, and strong and accurate processing ability.
For example, users of robotic vacuum cleaners know that they must remove all small furniture and other obstacles such as cables from the floor, because even the best robot vacuum cleaner cannot clean it on its own. The most difficult thing is that the robot has to operate in the presence of moving obstacles when people and pets are walking in close proximity.
While they seem straightforward for humans, many household tasks are too complex for robots. Industrial robots are excellent for repetitive operations as the robot’s movement can be pre-programmed. But household tasks are often unique and can be full of surprises that require the robot to constantly make decisions and change its course to perform tasks.
Think about cooking or cleaning the dishes. Within a few minutes of cooking, you might grab a frying pan, spoon, stove top handle, refrigerator door handle, eggs, and a bottle of cooking oil. To wash the pan, you usually hold and move it with one hand while scrubbing with the other, making sure to remove all cooked food and then rinsing off all the soap.
There has been a huge development in recent years using machine learning to train bots to make intelligent decisions when choosing and placing different objects, which means assimilating and moving objects from one place to another. However, being able to train robots to master all kinds of different kitchen appliances and household appliances will be another level of difficulty even for the best learning algorithms.
Not to mention that people’s homes often have stairs, narrow corridors, and high shelves. These hard-to-reach spaces limit the use of today’s mobile robots, which tend to use wheels or four legs. Human-like robots, which would more closely match the environments humans build and organize for themselves, have not yet been reliably used outside of laboratory settings.
The solution to the complexity of tasks is the construction of special-purpose robots, such as robotic vacuum cleaners or kitchen robots. It is likely that many different types of such devices will be developed in the near future, but general-purpose home robots are still a long way off.
According to Herid Assistant Professor of Mechanical and Aerospace Engineering at The Ohio State University. This article has been republished from Conversation Under Creative Commons License.