SafeAI: bringing 2.0 autonomy to mining
IM Editorial Director Paul Moore spoke to Brenton Welford, Vice President of Business Development at SafeAI, and David Prusinski, Director of Revenue, about its standalone mining solutions – from journey to now and what sets its system and strategy apart
SafeAI, based in California, is rapidly evolving from a young start-up founded in Silicon Valley in 2017 to an increasingly well-known company that is establishing itself globally in major mining markets and signing partnership agreements important with relevant companies. These include an agreement with tire giant OTR Goodyear to incorporate tire intelligence into the programming of autonomous heavy equipment vehicles, and a partnership with Japanese construction group Obayashi to create construction sites. autonomous, starting with a pilot program in the United States.
David Prusinski, SafeAI Revenue Director (left) and Brenton Welford, Vice President of Business Development (right)
In January 2021, SafeAI announced the opening of a new office in Perth, Australia, and welcomed three stand-alone mining specialists there, focused on leading its rapidly growing mining division in this part. of the world. Most recently, in June 2021, SafeAI secured $ 21 million in funding led by transformational technology investor Builders VC. The new capital will be used to advance the company’s open and interoperable autonomous technology through accelerated deployment and support, and will fuel global expansion to meet growing demand for autonomous heavy equipment.
Welford said I AM: “We are a rapidly growing venture capital-backed company, and self-sustaining mining is our central focus. We are considering transportation in mining and upgrading our solution not only on larger mining trucks, but also on smaller class trucks and other mining vehicles – this includes smaller trucks that could be used. for other tasks like storage work or earthmoving – in the case of Cat this may include the 773, 775 and 777 models as well as others of a similar class from Komatsu, Hitachi, etc. Then you have all the ADTs, water trucks, off-road dumpers with dump trailers and others that are unaffected by range. So we want to extend autonomous technology to all fleets beyond the primary overburden and ore fleets. This opens up autonomy to second-level miners and entrepreneurs (who may use smaller-class machines, or mixed fleets, as main fleets) as well as larger quarry operators – who do not both have been at the center of the concerns of the autonomy majors.
He points out that the autonomous trucks currently in service, both “new” and under renovation, still total less than 1,000 units – and are dominated by Caterpillar and Komatsu. The vast majority of these units to date, whether new or under renovation, also belonged to the 200t and less than 300t payload class and dominated by a few models such as the 793F and 930E. So why has autonomy not taken off in the trucks of the less than 100 to 200 t class, which represent the majority in terms of the number of operations?
Prusinski adds, “Part of this is the cost – and not just the cost of the vehicle or retrofit kit – it also relates to the investment in the network, the design of the mine and the additional infrastructure needed to operate. AHS reliably. We have to take into account the cost of autonomy versus the life of the mine and the time needed to pay off that initial capital investment. Many small mines also have a shorter lifespan, and contractors can only count on having their fleets in place for a few years. Our strategy at SafeAI is to fill this market gap.
Prusinski says SafeAI believes it can do this cost-effectively for these new vehicles and new markets by learning from the automotive market, where huge strides in Range 2.0 have been made. The idea is that in mining you don’t necessarily need to tie battery life to a huge network investment for small fleets – you can use a lot more on-board data processing. The technology, including sensors and compute platforms, is now much more advanced and enables more efficient stand-alone solutions, with a lower cost of entry and an overall lower total cost of ownership. “The mining and autonomous automotive journeys diverged in 2014, but there is now a lot to learn from what happened in the automotive industry where cost sensitivities are much greater. Our founder and CEO Bibhrajit Halder was part of Caterpillar’s initial autonomy program, before joining Ford where he worked on self-driving cars and Apple’s self-driving team, until he created SafeAI in Silicon Valley. Other SafeAI staff have similar relevant experience, so it’s understandable that the team is confident that they can bring some of this information back to the mining world.
So who is SafeAI’s target customer in mining and what has been done so far? Welford says: “We’ve only been going there for three years, but we’re already moving from proof of concept to the first on-site deployments of our task-specific autonomy platform at active mine sites. We have automated an ADT – the Caterpillar 725 in conjunction with the Japanese construction company and SafeAI partner Obayashi – as well as a Bobcat skid steer loader and several light vehicles.
Proof of concept tests took place at SafeAI’s own test site in Silicon Valley, a closed copper mine and a US quarry. In Australia, its first proof of concept will begin in 2021. Welford says: “Australia represents a huge market opportunity for us, not only for mid-level owner-operators, but also from the perspective of mining contractors. These customers demand a more flexible stand-alone solution than what is currently available, not only for the system itself, but also for the application model.
How is SafeAI’s offering different?
Fundamentally, to be able to operate in such a diverse market of large and small miners, across a range of equipment types and sizes, SafeAI’s system – both hardware and software – is designed to be interoperable with other systems. OEM, both on-board and in terms of FMS. Its application programming interfaces (APIs) are fully open and transparent. This gives a very flexible solution based on what the customer has in terms of machine types and models, but also whether they are running Wenco, Hexagon, Zyfra, Modular or another FMS – being FMS agnostic remains a big challenge for the customer. autonomy today, But it can be done. For small operators, SafeAI’s AI-based software can also be used to provide FMS functions such as machine coordination and truck dispatch assignment. Prusinski adds: “We are also an open system. Customers have full access to the data generated by the system. Mining companies are increasingly data driven, so providing this important data to help their business can only help the industry. “
The power of on-board processing and AI
The latest explosion in on-board processing capabilities, powered by GPUs and TPUs from companies like NVIDIA and Google, has unlocked huge potential for standalone use cases. SafeAI uses this advanced on-board processing power to enable its primary IP, which is the on-board stand-alone software, to meet multiple use cases that are critical to accelerating self-sufficiency in mining.
Welford said I AM: “An example is that in the systems in place, false positive obstacle detections are a known problem for effective stand-alone operations. Like SafeAI, these systems use truck-mounted sensors as part of their critical safety systems, and when these sensors detect an obstacle, such as an unexpected light vehicle ahead, the truck will stop. However, when the vehicle stops, it has no way of classifying the detected obstacle, so it waits to be cleared, in the field, often by a human operator who manually inspects the vehicle and determines the type and size. state of the obstacle. The problem is compounded by the number of other objects that the sensors also detect; these may be small clouds of dust, stones, birds or a number of other objects which do not pose a risk to the vehicle.
Prusinski adds, “The current systems of mining are in place and working, but that doesn’t mean things have to stand still; with AI, they could be even more productive. SafeAI’s technology means that the truck is able, under certain circumstances, to intervene when needed. Autonomy 2.0 integrates artificial intelligence, allowing it to detect, classify and track obstacles in unstructured environments, coupled with data fusion and advanced predictive algorithms that allow the system to detect and avoid obstacles in a safe and productive manner. In addition, the pit obstacle clearance method involves more people in the autonomous area, which is best avoided.
There are a few app caveats. The focus for SafeAI remains on transport variations i.e. transporting material from A to B, because once you get into other types of equipment it gets very complex. Take the example of wheel loaders – they do a lot of precision work in a variety of use cases – with wheel loaders around inventory for example. You need 100% utilization in one or two of these cases to make automation worth it. Utilization rates also determine whether transport units are worth automating. Welford comments, “We are working with miners and smaller contractors to understand how much of their fleet they need to automate and the level of equipment usage required to make a meaningful return on their investment. This could make the transition to autonomy more difficult for certain operations, for example those operating for more than 12 hours a day. “
So where do we stand here for SafeAI? Welford concludes, “We are based in the US, we are aggressively hiring to expand our Perth office and have a presence in Canada; North America and Australia therefore remain the key markets for our mining focus for now. Of course, there is also a large construction and mining market in Asia, and given SafeAI’s strategic investor, Obayashi, who is also on our radar. But we’re not limited to those regions – our goal is to get as many initial sites as possible, with a variety of use cases, operator types, mine types, and vehicle types. In this way, we can show the real power but also the flexibility of our system compared to what is currently commercially available. “