Occam's Razor is the principle that a solution to a problem should be as simple as possible -- that we prefer the solution which depends on the smallest set of elements.
As engineers, startup founders, and tech enthusiasts in general, we often lose sight of this principle. When we look at the kinds of problems that robotics aims to address, I often observe us jumping ahead to "AGI level" fully autonomous humaoid robots, hyper-sophisticated robots that cost hundreds of thousands of dollars and perform highly complex tasks. There is definitely a place for this, but my thoughts are that we should first capture the low hanging fruits of automation.
Finding Low Hanging Fruits
I often find myself asking the question "what would it take for me to want to do this job" about most jobs. For some, like building interesting products, the answer is nothing, I'd be doing that anyways. However for some jobs, like deep sea exploration, mining, search and rescue, I feel like I would be putting my life at unnecessary risk -- because we have the technology to make these jobs a lot safer.
Some dangerous jobs are hard to automate, but it may be quite doable to make them a lot safer for human operators.
A principled approach to problem-finding
The approach we aim to take in this project is to first establish a list of specific dangerous tasks - the ones that cause the most deaths and injuries. With this list in mind, we will also score them by how difficult it would be to mitigate the human risk. After this process, we will be able to sort by the danger/difficulty ratio, and solve these problems in that order.
The key distinction here is that we do not aim for automation, only for safety. Similarly to successful companies like SawStop.
An example we chose to pursue as the first target of practical robotics is the forestry industry. Forestry is a huge industry in BC, and one of the most dangerous.
Remote controlled robots with simple functions can greatly reduce the risk in this industry, even something as simple as a tripod mounted 3dof axe, which can deal the final blow to a tree from a distance.
Project Stauts
Currently, we are aiming to increase the number of early customers for the forestry project, and to scope out other focus areas with strong market potential. If you're interested in collaborating on this project, do not hesitate to reach out to hellorobotics@thebigsasha.com