“I’ve only had one job in my entire life: this one. We’ve been promised robots in books and movies, but I haven’t seen any. I founded iRobot after graduating from MIT with the goal of advancing the robotics industry.” He succeeded, at least partially. Colin Angle (Concord, USA, 56 years old) has been making robots of all kinds for more than 30 years. The first steps were taken with the Defense Ministry, to which it has sold thousands of tracked robots to defuse anti-personnel mines. These are reconnaissance robots that traveled the ruins of the Twin Towers in search of survivors after the attacks.
However, he owed his great success to the Roomba robot vacuum cleaners. Almost a pop icon, you opened up the domestic robotics market, which this engineer says will be “huge” in a few years. “The original Roomba algorithm is the same one used by the robots we developed for the Pentagon, but with a built-in vacuum cleaner,” says Angle, sitting cross-footed on a sofa and speaking slowly.
iRobot devices have multiple sensors and powerful software that allow them to move with reasonable precision. They vacuum and wipe independently and quite efficiently. They can detect whether we are at home or not (by searching for our devices via Bluetooth) and activate them when they are not disturbing. Its technology has not gone unnoticed among the giants of the industry. Amazon announced its intention to acquire the company in August last year. The operation still has to be approved by US regulators, which will examine whether it violates antitrust laws. The European Commission launched its own inquiry into the matter this summer. “We believe that Amazon is aligned with a very long-term vision and that this can give iRobot the opportunity to innovate faster than before,” he says. Angle visits EL PAÍS in Madrid during his European tour to introduce the latest Roomba models.
Questions. What is new in the latest models compared to previous models?
Answer. We now have the best robot vacuum on the market, the J9+, and the best robot vacuum and mop, the J9+ Combo. We have achieved that the robot can clean for 60 days without having to touch it. Also with the mopping robot: While the water previously had to be changed after every expedition, this maintenance now takes place automatically at the emptying, refilling and loading station. The vacuum cleaner is now more powerful and we have a system called Dirt Detective that can clean any room to your liking.
The Urbie robot, one of the first robots launched by iRobot, was designed for military use and rescue missions. NASA/JPL/CALTECH
Q What does it refer to?
R We have integrated software that creates a map of the house and the different rooms. On this you can see what is in each room, how dirty it has been in the past and other considerations that will help you know what is happening in every corner of the house. For example, that the dining room should be cleaned more often than the hallway and always after meals, or that it is better to clean the bathroom after the kitchen and not the other way around. Creates specific cleaning schedules based on floor type and accumulated debris, and predicts priority areas.
Q How do you imagine robot cleaning in ten years?
R The intelligence that makes this robot so great can be applied to even more things. Why can’t we tell other devices in the house where they are and what we need? For example, air purifiers are often set to low power because they make a lot of noise when they are ineffective. Why can’t the device detect when it’s a good time to run loud without disturbing?
Q Is your idea to integrate Roomba robots into more home appliances?
R The concept of a smart home is very confusing today. There are some good ideas, but there is a lack of intelligence. I think we need to look at the home more as a robot, as a system. And in 10 years, iRobot will certainly want to do more than just vacuum and mop the floor.
Colin Angle poses next to the Roomba J9+, whose new docking station allows you to change the water for scrubbing and empty the waste tank, giving it 60 days of autonomy.Claudio Álvarez
Q Amazon has smart home devices such as the Echo voice assistant, Ring doorbells and cameras or the Astro robot. Would they fit into this plan?
R Amazon has a great catalog and we have a good product and technology. It is very important to him to fascinate his customers. They design products for many areas of the home. Our visions agree, yes.
Q Will we end up talking to the vacuum cleaner?
R Naturally. We’re not working on that, but Google, Apple and Amazon are developing conversation models for their smart speakers. At the moment, it’s difficult to control a Roomba with your voice because you have to give the commands [a Alexa u otros asistentes] with a very specific syntax, and that can be frustrating. But if we’re able to develop a conversational interface where I can say, “Hey, Roomba, can you clean up the kitchen when I’m done eating?” and Roomba tells me it’s okay, then everything will work smoothly. We will open a door to a new way of interacting with technology.
We know that there is a rectangular item called a kitchen in your house, but an online sales portal knows the last 78 things you have bought
Q The information that Roombas robots collect from homes is very confidential. How do you protect this data? Are these vacuum cleaners safe?
R The information collected by the robots does not leave the devices, so the images and plans of the house remain in the vacuum cleaner unless the user specifies otherwise. The data transferred to the user’s phone is end-to-end encrypted and therefore not vulnerable to theft. The next level of security is that the data we manage is stored in a way that is only useful to the robot. This means that if you managed to bypass all of our cybersecurity measures and get to the blueprints of a house, you would be disappointed by what you would see: a series of polygonal shapes labeled “kitchen” or “bedroom.” , because the robot doesn’t care about any further information. We do not store people’s private information as it is sensitive and we do not need it to function. I want customers to make purchasing decisions based on the privacy the products offer.
Q iRobot has access to confidential information about houses: their physical layout. The rest of the tech companies don’t have this data.
R We know there’s a rectangular object called a kitchen in your house, but an online sales portal knows the last 78 things you’ve bought and can draw more comprehensive conclusions about your behavior than we can. I’m not sure we can contribute much to these predictions with what we know. And I don’t want it either. Our commitment to data protection is profound. We will never sell customer data, we will make money in other ways. My long-term vision is to help seniors live more independently. I don’t want us to create a data vacuum. Actually, I’m not interested in this data. If you buy a Roomba, I want you to never buy a cleaning robot again.
I don’t want Roomba to be a data vacuum cleaner
Q However, when Amazon announced its intention to acquire iRobot, many thought the company was after their data.
R I know. I think that Amazon, like the rest of retailers, has much better ways of getting information from users than through our robots. We don’t provide them with a tool that improves on what they already have in this regard. I don’t think they want to buy iRobot because of the data we have because it’s irrelevant.
Q Despite the security measures you describe, private pictures of Roomba users taken with their vacuum cleaners circulated on social media last year.
R They do not come from consumer robots. This couldn’t have happened with a commercial Roomba. The images belonged to users who were testing models in development and consented to iRobot recording and collecting images. They were leaked by employees of an external company [Scale AI] is dedicated to image annotation for machine learning models [etiquetar imágenes manualmente para que la máquina asocie una serie de píxeles a una idea]. It was unfortunate, but it was not a consumer problem, but rather a development problem. The company is now facing legal problems. We will ensure that something like this can never happen again.
Q Are you still working with image annotation companies?
R We continue to work intensively on computer vision. We’ve added new protections for image annotation efforts. For annotation, we use automated tools that give us more control over the process and do not have to deal with third parties as before.
You can follow EL PAÍS technology on Facebook and Twitter or sign up here to receive our weekly newsletter.
Subscribe to continue reading
Read without limits