When one talks of ‘future compression’, perhaps the reader recalls Moore’s law or maybe you imagine the future drawing exponentially nearer. If you’re in the latter group, you’re close to the mark. The widespread deployment of AI, Saudi Arabia granting citizenship to a robot(?!) and Davos’s asteroid mining executive (?!?!) suggests the future is now- the 4th Industrial Revolution is upon us. Peter Diamandis, co-founder of Planetary Resources, and Davos’s aforementioned asteroid mining executive is adamant that space mining will be part of this future. Larry Page and Richard Branson, both investors in Planetary Resources, apparently share this vision. But, before we make it to the great beyond, our space-savvy entrepreneurs need an ERP to get them off-the-ground and traditional systems might not be up to the challenge.
A very brief introduction to space mining
Space mining does what it says on the tin. Deep Space Industries and Planetary Resources, resident to the wider space industry, are trail-blazers in the industry that plans to, as J.S. Lewis eloquently put it, ‘mine the sky’. The rewards are certainly impressive: in July, an asteroid with a $3.5 trillion platinum core passed near earth. Jakhu, Pelton and Nyampong’s seminal text, ‘Space Mining and Its Regulation’ outlines the three criteria asteroids need to meet to be considered. They must: (i) be reasonably large, being more than 100 meters in diameter, (ii) be traveling along an approachable orbital path, (iii) have a high percentage of high-value natural resources. These near-Earth asteroids (NEAs) are then ear-marked as candidate asteroids.
Not so fast. Whilst, Goldman Sachs has highlighted that “financial and technological barriers” to asteroid mining are “far lower” than the psychological hurdles we will need to clear, space mining is not easy. In fact, it is rocket science, and then some.
The major challenges facing space mining ventures
Space mining companies need to get off the ground- first financially and then literally.Google’s LunarX challenge offers companies the opportunity to pick-up $20 million if their robot can complete a series of tasks on the moon. There are other ways to raise funds also. Ingeniously, Planetary Resources planned to raise money for their ARKYD satellite by enabling backers on Kickstarter to have their ‘space selfie’ taken. Unfortunately, the campaign came to nothing. Currently, these ventures mainly rely on capital raised from investors but they are beginning to hold their own. Insofar, it is apparent that these canny companies can raise significant capital through winning satellite-focused commercial contracts where they can boost the bottom-line and learn at the same time. In early 2016, HawkEye360 chose Deep Space Industries as “the satellite provider for its Pathfinder small satellite mini-constellation” which will collect information to monitor global transport. Insofar, I have assumed that our companies have been operating on fairly standard ERP system, either built by themselves or imported from a software company. As these enterprises expand into the great beyond, ERP is going to get messy and might not be needed at all. In the following paragraphs, I will outline the basic space mining process (excluding manufacturing and return to Earth for the sake of brevity), and then consider whether ERP is suitable for space mining at all.
Identifying candidate asteroids
AsteroidZoo, the result of the combined efforts of Zooniverse and Planetary Resources, is a citizen-based project that has identified thousands of asteroids. Planetary Resources can then classify these asteroids by their diameter, time to reach, type of orbit and the velocity needed to reach from Low Earth Orbit (LEO). Public databases and private databases will be vital when these companies come to decide which asteroids to target.
Reviewing candidate asteroids
Planetary Resources suggest that as early as 2020 they will be touching down on the candidate asteroids. Before we proceed it must be noted that Planetary Resources (and other space mining companies) plan to proceed with their operations in two stages: (i) targetting water-rich asteroids, and extracting the resource, (ii) targetting metal-rich asteroids, and extracting the resource.
Beginning the process, Arkyd-6, a broadband imager, will identify candidate asteroid’s with a high water content. In space, water is useful as-as propellant, a life-support and for hydroponics. Being able to covert super-heated water into fuel is useful- it means that craft can spend extended periods away from Earth. Once the candidate asteroids have been identified, the start-up’s Arkyd-301 probes. The data collection “will include global hydration mapping and subsurface extraction ... to determine the quantity of water and the value of water resources available.” At a later date, space mining companies will target asteroids rich in metal.
Reaching the asteroid
There are a number of transport technologies that can be deployed to reach the candidate NEA. These are concisely outlined in ‘Space Mining and Its Regulation’. Solid and liquid chemical-fueled launchers are standard space-flight fuel. For example, Space X’s Merlin engine utilises a combination of refined kerosene (RP-1) and liquid oxygen. This tech is very useful for getting out of LEO, but less so for delivering significant thrust over a long period of time.
Ion propulsion systems are used to provide thrust over a longer period of time. Planetary Resource’s information page on Arkyd-301 suggests the probes will be launched into space with chemical fuel rockets and then continue their journey to the asteroids using ion propulsion systems. NASA’s Dawn craft used ion propulsion to visit the dwarf planets Vesta and Ceres. Dawn has a fuel efficiency 10-12 greater than chemical thrusters. But, it does take 10 000 hours to accelerate the spacecraft so speed necessary to reach the asteroid belt. However, one can only envision that this time will be reduced in the future.
Other more long-term options for transportation include nuclear-fuelled propulsion (in some instances, engines could be powered by helium-3 mined from the Moon), mass-driver systems (electromagnetic cannons for launching materials off low-gravity bodies) and space elevator systems. The later is becoming increasingly unlikely thanks to SpaceX and Blue Origin’s efforts to build reusable rockets.
After overcoming the logistical nightmare of getting to the asteroid itself, it seems expedient that one would then set about mining it. Minning an asteroid requires the successful deployment of Space Robotic Mining Systems (SRMS), and Jakhu, Pelton, and Nyampong have identified challenges faced by these systems. With our focus on ERP, some are of note to us:
Using less energy when mining
Improved software and AI to allow for teleoperation and automation
Using smart 3D printers to manufacture using locally acquired materials
Upgradability and modularity so broke or obsolete tech can be replaced
They point towards futuristic space mining techniques, with electromagnetic weapons and “large amplitude sonic frequency generators” not crossed-off the roster just yet.
Are ERP systems suitable for space mining?
ERP is the “central nervous system” of a business, giving corporations the mans to become self-aware as to their activities.
In a recent Reddit AMA, a SpaceX employee provided a lucid insight into how SpaceX’s Enterprise Information System (EIS) functions. The EIS team have developed their own internal web application. It is used by all employees, including those who are “creating purchase orders… filling our part inventory, engineers creating designs and work orders with those parts, technicians on the floor clocking in and seeing what today’s work will be per those designs… and literally everything in between.” Not only is the system hyper-efficient, but because the system is produced by talent within the firm, the TCO is lower and the system can be adapted to meet the firm’s needs exactly.
One feels that space mining ventures should, if costs permit, develop their own ERP as opposed to relying on systems such as SAP Business One or SAP ERP. SAP systems are no flexible enough to deal with the complexities of a space mining operation. And, as is highlighted above, the TCO is just too high for companies having to devote the majority of their resources to developing their SRMS and transportation technologies.
Space mining companies could use their own ERP whilst there operations are Earth-based, but one feels that ERP systems will have limited application in space. If space mining companies don’t use traditional ERP methods, what will they use? Since we are yet to actually begin mining in space, the answer to this question will be largely theoretical. As outlined above, there are four major hurdles that need to be overcome if we are to mine the sky: identifying candidate asteroids, reviewing candidate asteroids, reaching the candidate asteroid and excavating the candidate asteroid.
In producing the following outline, I have assumed that for efficiency the undertaking will be as automated as possible. On a side note, If you’re interested in learning about the role of AI and robotics in space mining then offworld.ai is a great place to start.
AI is not only the future of ERP, but will go beyond ERP. Man Group, a $96bn hedge fund, allows AI to manage some of its portfolios. Could AI manage a space mining operation? Let’s not kid ourselves. Whilst managing a hedge fund is certainly complex, it is not comparable to the labyrinthine details of a space mining operation. But, this does not exclude the possibility that AI will be managing future space mining operations.
Put simply, AI is better than ERP because it can teach itself. The effectiveness of an ERP system is limited by the capacity of the humans using the system: ERP sets-out the rules, and the workers play by those rules. For AI, this is not the case- it is, as far as its operator allows, the player and referee both setting the rules and playing by them. Moreover, ERP systems need have updates rolled-out as the system adapts to a general change. Conversely, AI will most likely be able to update itself with a near instantaneous quality. Beneficially, the update will be specific to the company’s needs as opposed to a general update covering many companies and industries.
Now, let us examine how, with an advanced AI, Planetary Resoucrces Inc. might manage their space mining operations. Our hypothetical AI would receive information from observatories
—perhaps automated— to help it select candidate asteroids. The AI would then direct the Arkyd-6 model(s) to review these asteroids, with the candidate asteroids be listed by the most profitable to the least profitable (with data from future global market trends and the actual costs of the mining process taken into account). In a matter of minutes, the launch order for the appropriately positioned Arkyd-301 probes would then be executed and the mining process would begin. All these tasks above could be performed by a human, albeit in a much less efficient manner.
Of course, my conception of such an operation is limited by my own imagination and lack of data. It is likely that the AI will account for the data of all asteroids and all operations when making operational decisions. For example, the AI might go right on ahead and completely redesign the system and the technologies used by our entrepreneur’s companies. The AI might discover a more cost-efficient way to manufacture fuels in space or invent an entirely new fuel which would cut R&D and engineers on the ground out of the equation entirely.
I foresee that some might rail against me- what if profitability is not the outcome we desire? If it is not, then there is no problem. Unlike ERP systems, AI can run inside simulations that account for all the real world data. The company would then run simulations focused on their desired end, let us say sustainable space mining for example. This module could then be deployed in the real world.
In last analysis, the future is here now- or at worse, just around the corner. In the space mining industry, at least, I foresee that ERP will be of little use in managing the involuted intricacies of commercial space mining operations. The foreseeable transformation in the space mining industry is part of the larger, long-term, alteration from ERP to AI.