How to create autonomous vehicles from scratch (and existing models) with Mel Torrie

author avatar

17 Oct, 2019

How to create autonomous vehicles from scratch (and existing models) with Mel Torrie

Mel Torrie, is the founder and CEO of ASI, Autonomous Solutions Inc. He talks about how ASI develops a diversified portfolio of vehicle automation systems across multiple industries.

Mel Torrie, is the founder and CEO of ASI, Autonomous Solutions Inc. He talks about how ASI develops a diversified portfolio of vehicle automation systems across multiple industries. 

ASI often collaborates with OEM’s, leveraging already existing vehicles and adding their automation expertise. Mel also talks about his goal of growing a great company for the long haul and shares how they survived the 2008 crisis. View the interview to find out how the Google self-driving car has helped ASI gain traction! 

Interview: How to create autonomous vehicles from scratch (and existing models) with Mel Torrie    

Per Sjöborg, host of the Robots In Depth podcast, interviews Mel Torrie, the founder and CEO of ASI, Autonomous Solutions Inc. Below is a transcript of the interview. 




Per Sjöborg: Welcome to the podcast version of Robots in Depth Episode 7 with Mel Torrie in cooperation with Wevolver. Robots in Depth is supported by Aptomica. Visit Aptomica.com to connect. You will find all past episodes and more in RobotsInDepth.com. Welcome to Robots in Depth. Today I'm honored to have Mel Torrie from ASI. We’re going to talk about his ability to automate any vehicle I would think. That’s what you do. You take an existing vehicle and you make it automated, fully automated or partially automated.

Developing autonomous vehicles from existing vehicles

Mel: Yes, I think it's more fun to make from scratch type platforms and we do that but most of the demand has already vehicle solutions in place. It's we've got a big mining truck, we've got a farm tractor, we've got a factory robot and we just need it automated to be more capable and less labor intensive.

Per: You started this company from scratch. Could you tell me about the start-up phase when you got into robotics and started the company?

Mel: About 1997-98 I wrote a paper on some of the robotics we were doing in the Department of Energy for the government while I was at the University and in a robotics lab there. John Deere saw the paper. They engaged us and we started to do some safety analysis for robotic tractors. We had a little mannequin with a three-year-old sized person on it come running at the tractor and we had to prove that it could be safe. Once we prove the safety John Deere said hey, let's go and go to production. ASI, myself and some of the university professors and some students spun out the company, started it up and we got some John Deere projects to begin with and won some government contracts and then diversified into mining and industrial cleaning and factory warehouse stuff.

Per: Industrial cleaning, could you talk a bit more about that?

Mel: We're trying to use our building blocks, our software platform across all of these vehicle related industries and so industrial cleaning is another one of those great applications. Whether it's floor sweeping in the Home Depot, Walmarts up through the factory warehouses to parking lots sweeping, that kind of thing.

Per: Are you operating mostly in the night time then when customers are not present?

Mel: This just started a couple of months ago so we're in the product development phase with an OEM that sells the standard machines. 

Per: Again you're adding your technology to an already existing system because those are complicated enough on their own.

Mel: Getting the cleaning technology right is definitely a huge challenge and so instead of trying to build up a floor scrubber from scratch and try to compete with people who've been doing it for years we partner with them. Depending on the industry if there's already OEMs established that are very hard to compete against then we'll partner with them. They have the distribution and service all in place.

Per: You're not going to build a better tractor than John Deere. It’s hard.

Mel: Yes and try to compete with their 10,000 dealerships. There it makes more sense to partner and then in other industries it makes sense more for us to go out there and introduce the product in the market.

Per: How did the project come about?

Mel: The demand heavily is on contractors. The Walmarts of the world are typically not buying their own cleaner and staffing. They are getting a contractor to come in and the contractors are the ones that are really pushing for this technology.

Per: The contractor sees that they could provide a better service and a cheaper service to their customers.

Mel: It's definitely based on doing it better, consistent work. I was surprised that warranties on carpet and floor tiles are based on or they're impacted heavily on how often they're cleaned and you can void your warranty by not cleaning them effectively. There’s definitely huge reasons for getting it done very thoroughly instead of just I see dirt there I go clean there and getting it done much faster but not doing a thorough job. Also the labor challenges, low skilled labor and being able to expand so you can have one person more skilled, educate them and have them manage more robots. Maybe they're managing all of the Walmarts in Chicago versus one person per Walmart.

Per: Then they'll address the problems when the robot has an issue they can't solve themselves the person goes in and takes care of that. Just like we all use tools just that now it's an automated floor cleaner rather than a hammer. It’s just more advanced tool.

Mel: There's other things they can do that the robots can't whether it's cleaning the windows, cleaning the bathroom and then they can go on service that robot.

Per: They do both. They manage the robot doing what the robot is good at and then they do the rest and then they move on to the next venue. It’s just a smarter tool.

Mel: Yes and they're getting far more use out of the equipment because they're able to move from Walmart to Walmart and move much faster and more effectively.

Per: That's a very interesting project. You started out with doing the tractors. Did you then go on to do more agricultural automation?

Agricultural robots and automation

Mel: The first year we really pushed on the orchard vineyard tractors for doing spraying of dangerous chemicals. We went and did a golf course mower for free in our overtime. We said hey, Mr. John Deere if you'll send us a golf course mower we think that's a great application. We will automate it for you and demonstrate the capabilities. Brilliant entrepreneur, risk-taker that usually loses money on those kinds of things. We worked overnight, overtime Saturdays and got together a golf course mower and we were able to convince them of the practicality. Then we got some projects that way and then someone from mining saw us, a big mining company. They came to us and then we started developing autonomous bulldozers and mining trucks.

Per: The safety and security there must be just paramount. I mean they're operating in a dangerous environment and they're operating some very heavy machinery, very powerful machineries.

Mel: Yes, 420 tons. It will run over a pickup truck or a house without feeling it.

Per: That's two blue whales. That’s just unimaginably heavier. I guess having a good track record that the partner trusts you to do a good job is the long-term relationship you have with both earlier customers and the new customers has to be just so important.

Mel: That resume has been critical for us as a small business to be able to grow without investors. You thrill those big customers. You’re able to stay with John Deere for 11 years, we must have done something right. That resume is just critical for us to grow into those other markets.

Per: I know you call yourself a small business. I'm not sure everyone would agree. You’re pretty big but do you see yourself running this business in a different way. Could you talk a bit about how you see running your business?

Growing a robotics company

Mel:  I definitely got depressed after reading Steve Jobs biography that I just wasn't the jerk that I needed to be to take this company to the next level. I'm working on that but I think trying to do it where you're not forced to take an exit through investors that 3 to 5 year kind of turn whether it's an acquisition or going public. I want to grow a great company over the long haul because what else would I do? There’s not much more fun than driving a 400 ton remote-controlled toy.

Per: That's a remote-controlled car.

Mel: I don't want to get out of the business so how do you do that? How do you stay in control, take care of these people that bought into your story and convince them to come? It’s really been a bootstrapping approach with large OEMs, original equipment manufacturers and large end users that have the money to help fund some of these early prototypes. Then we built the software platform that allows us to quickly leverage the technology and adapt it to each of these markets. Then putting people over each of those markets so it's like their own business unit that I am measured by how well I do in agriculture. We call it diversified focus but each of those markets gets a team that's focused and a leader that's focused on that market and knows that market.

Per: But still can benefit from being a part of the larger company with fundamental technology and developed in the company. Also the skill and experience of everyone throughout the company.

Mel: As we look at venture capitalists, they're the smartest in business that have been through it and grown companies and yet they still bet across ten companies because they know only 20% of those are going to hit. I have to have that diversity to survive in these rough times.

Per: You've been through the 2008 crisis.

Mel: I don't want to talk about it. It was rough.

Per: Because for new entrepreneurs out there this is very important. It’s not always going to be up and up and up. There’s going to be other days too and for them to hear your story I think it's very beneficial.

Autonomous vehicles in mining, construction and industrial cleaning

Mel: It was definitely a surprise. We had diversified across mining and agriculture and construction and industrial cleaning, all of those and they all pretty well crashed within three weeks. We had way too much work, too much money and then within three weeks we were in deep trouble. The government work was really the only stuff that kept going. All the other industries the commodity prices dropped and ag dropped and the commodity prices in mining was out. One by one they fell and it was fast. The diversity definitely is what kept us in business. If we were in mining as I was told to do that I was an idiot for not throwing all my eggs in one basket and doing that one better I wouldn't be here today. Diversity was important and being diversified across more than just commodities based. That was critical in surviving. I think running lean obviously was critical. I wasn't paying myself $200,000 and all of these other people. We were pushing for really getting the products out the door and being successful and so we were fairly lean and that helped us overcome.

Per: That also allows you to build this long-term relationship with different partners like John Deere because you survived one crisis and hopefully there's going to be a number of years to the next crisis. The world is really built by survivors because the other ones are not here anymore. When also your partners know that if there is rough times ahead they're going to be there anyway. We can trust them. That has to be a huge benefit when you're trying to acquire new partners or do new projects with your existing partners.

Mel: Definitely especially at mining where the swings are so heavy that plenty of businesses went out in 2008. For them to see that diversity and the other exciting thing they see is wow you're bringing in this technology from ag and this technology from mining and this technology from automotive and so that story that diversity we're going to be around a long time and a great resume of customers. We’re bringing in technology from all those other industries you don't have to pay for has definitely been a business model that's worked.

Per: I mean you've been in the robotics industry. You’ve been there doing the real hard work, your field robotics. I mean robotics is growing.

Mel: It's exciting that people are calling that wouldn't talk to me in 2009. The investors, it seems like robotics has jumped onto that top five things that's really going to take off and so we've really had investor after investor calling and calling.

Per: Also new customers from new segments, are you getting calls from industries that you didn't do before?

Mel: Definitely, right now we've got too much work. We’ve got that challenge of hire, hire, hire, all my people want more staff but you look at that next recession and you have to be careful. It’s a balance of hitting these opportunity windows and not having so much staff that a layoff is inevitable.

Per: You're screwed when it comes to the next test.

Mel: Yes so very exciting time. We are slammed. We are hiring ten a month and it's tough to find great people that we can bring in. It's a wonderful problem to have. We’re excited to be on that part of the curve right now.

New exciting applications in robotics

Per: Robotics is growing. Where do you think we're going to see the new exciting applications that are going to be visible to the general public? I mean not to people in mining, I mean you have a very broad view of robotics. You know about sensors and software and hardware. What are the general public going to see coming from robotics over the coming years?

Mel: I'm biased because I'm in more of the service industry so I think you're going to see street sweepers going past you without people. You’re going to see the parking lots getting cleaned. You’re going to see the tractors driving in the field as you go past and double take. I think that's going to be become very prevalent. You’re going to see the floors in the mall being cleaned and you're going to be walking past robots on a regular basis that way. I think in the public in my sphere that's what you're going to be seeing fairly quickly.

Per: This also will of course get people used to robotics when they see the cleaning machine in the mall, when they see the tractor they see that tasks can be automated. Then they will come back and request more automation because they're used to it. They’re not afraid of it anymore.

Mel: Google has helped us a lot in that way because of the Google Car. That press of wow, if they can do the car then surely they can do my tractor application or my mining application. We get that feedback. We had a mining company in the oil sands come to us last month and say surely my uncle's doing this on his farm or Google's doing this on TV, surely you can make my mining truck do this.

Per: Then you have to kind of tone it down.

Mel: There's definitely some unrealistic expectations of the press that what they're saying is possible but I think both the press that they've put out there as well as the legislation that they've been working on. Surely if they're legalizing driving cars on the highway surely we can get something on an orchard and a vineyard that's already secluded and it's got fences around it. They are spraying dangerous chemicals and they already keep people away surely we can legalize that. The Google car and the press that they've put behind that has been wonderful for all our industries.

Per: I've seen many of these things coming together like the public awareness, legal frameworks, safety regulations. We have a new ISO standard for safety when you cooperate we're robots. Lots of these things coming together and we'll see amazing things coming out of robotics.  I am sure you're going to be the one to bring it to us. I'm sure I'm going to be talking to you in a few years’ time and we're going to be talking about how we see the cleaning robot outside here in the commercial venue where we are. It’s going to be one of yours.

Mel: That's the plan.

Per: Thank you very much for being part of it and sharing your views on this and it's very important to us.

Mel: Thank you for having me.

Per: I hope you liked this episode of the podcast version of Robots in Depth. This episode is produced together with Wevolver. Wevolver is a platform and community providing engineers informative content to help them innovate. It is how engineers stay cutting edge. Aptomica is the founding sponsor for Robots in Depth. Aptomica runs anything in modular robotics. Dream, rent, build. Visit Aptomica.com to connect. I am your host Per Sjöbor. Until the next episode thank you for listening.

END OF TRANSCRIPT


Follow the podcast on Apple, and Spotify.


The Wevolver Robots in Depth podcast is published multiple times per week. If you, or your company is interested in supporting this podcast and reach an audience of professional engineers and engineering students, please contact us at richard [at] wevolver.com

 

More by Per Sjoborg

I interview CEOs, founders and CTOs from successful and up and coming robotics companies, as well as researchers, investors and thought leaders within robotics and AI. You'll learn about the latest technologies that are being developed at innovative robotics companies, the lessons learned by incredi...

Wevolver 2022