The robots are coming, and they’re getting smaller, smarter and cheaper.

While today, businesses may own just a single drone, one day, large parts of whole industries could be overseen by a new generation of robots. But, how will groups of these robots perform useful tasks like collecting crop data or organizing around a common goal?

That's one role for the blockchain proposed in a new white paper from MIT Media Lab research affiliate Eduardo Castelló Ferrer. The eight-year robotics veteran believes that by using the blockchain to distribute information, whole armies of robots could solve problems and accomplish tasks more efficiently.

The idea is that in robotic swarms, each robot follows basic rules, drawing inspiration from creatures like ants and fish that often clump together. These small rules then add up to collective behaviors, such as distributed sensing or search-and-rescue missions, that emerge as a result of the interactions between robots.

So far, these ideas haven’t been deployed on a large-scale. But researchers have high hopes for use cases such as so-called "precision farming," where fleets of drones could be used to inspect crops and paint a more granular picture for farmers.

As researchers are moving forward with this futuristic idea, they’re facing many security and logistical problems that have prevented robot swarms from moving from research labs to the real world. But, Ferrer thinks that the blockchain could lead to "serious progress" in swarm robotics.

The white paper explains:

Applying bitcoin to robotics has been theorized before for autonomous networks of driverless cars or drones that would deliver packages.

But this is different; the white paper sketches a blockchain-based system where robotic "nodes" organize in a secure, distributed way. One potential role for the blockchain is to help the robotic groups to come to agreement about a decision without a central authority.

It outlines a model where the robotic swarms use the blockchain by serving as nodes in a network and "encapsulating their transactions in blocks". Blockchain-based applications described in the white paper include secure communications between robots, distributed decision-making, behavior differentiation and new business models.

With the declining costs of robotics, Ferrer argues that possibilities abound for robotic swarms.

"Hardware is getting better. We can now make thousands of robots for research purposes, but also for industrial purposes," Ferrer said.

But one problem with current distributed algorithms, as the white paper describes, is that there are tradeoffs between speed and accuracy.

The idea is to leverage the blockchain for the best of both worlds.

"Blockchain is an outstanding technology for ensuring that all participants in a decentralized network share an identical view of the world," the white paper explains.

Robots might come to agreement about which direction to move or which shape to transform into by voting on the blockchain, a use case that political groups, such as Australia’s Flux Party, have been experimenting with.

Say that a group of sensor drones are debating whether an image is of a cup or a two faces. (This example from the white paper uses the classic rubin vase visualization). One of the robots will detect the need to make a decision and issue a vote using a special transaction creating two addresses that represent each choice: a cup or two faces.

The outcome is determined by majority rule. Each robot casts a vote in the form of a transaction sent to the address that represents their choice. Since the blockchain is public for all the robots to view, they can quickly verify the result of the vote. The robots repeat this process for each decision the robotic swarm needs to make.

Each vote, of course, would need to happen in a timely manner, which is one of the issues with blockchain technology as it currently stands, as public blockchains often require a minimum amount of time to ensure that the transactions are secure.

Robotic swarms will also need to jump from behavior to behavior to accomplish their objectives, according to the white paper.

To accomplish this, the groups of robots could use interoperable blockchains to swap control algorithms in a process the white paper calls "behavior differentiation". (Developers are currently trying to make this possible with ideas like pegged sidechains).

So, robotic swarms might switch behaviors when they move to another sidechain, as sketched out in the white paper. One blockchain might rely on decentralized control based on round-robin mining, for example, while another would have a "leader-follower" control mechanism.

Another differentiator might be private and public blockchains.

One sidechain might be a private company blockchain used for testing purposes and filled with company information, but it could be pegged to a public one. (Ferrer offers MultiChain, the build-your-own-blockchain company, as an example).

Those aren’t the only ideas described in the white paper. Blockchains could help create a system of trust to combat the rise of robots with malicious intent within the group, and these robotic swarms could also form Internet of Things markets, collecting information together and sending it to interested users.

But blockchains have some fundamental limitations, that Ferrer believes could block widespread use of blockchain-based swarm of robots.

For example, the 10-minute transaction confirmation time (roughly) of bitcoin stands in the way of quick voting on decisions. Robotic swarms probably don’t have 10 minutes to determine which way to move if they’re 30 feet from a potential obstacle.

Furthermore, if the bitcoin blockchain continues to grow, it might be too burdensome for each robot to carry or fly around with a copy of the full ledger.

Still, the white paper is optimistic about future of combining blockchains and robotic swarms:

Robot child image via Shutterstock