 |
| This is a deformable mobile robot, but NOT the one mentioned in the paper. |
The research paper is “Design and Locomotion Analysis of a Novel Deformable Mobile Robot with Two Spatial Reconfigurable Platforms and Three Kinematic Chains." You don't need to worry about the content of the paper per se; the abstract (available at that link) is good enough. Indeed, even the title of the paper is enough to make my point.
The authors report on a new design of deformable mobile robot. The design, novel though it may be, is not a whole new class of product. It is solidly based on existing technologies, physical principles, and even structure (though those parts have been reshaped significantly - rather like a whale's fin and a human's arm significantly reshaped versions of the same basic structure).
The robot can be thought of as a significant mutation on existing designs of deformable mobile robots, and all the other robots on which this new one is based are its genetic parents.
The researchers intend the performance of their new design to exceed that of other robots. While there is no intention in natural evolution, we don't really need to worry about that here because, depending on context, lots if not most new products fail [1-4]. And certainly if one only looks at early development projects - like this deformable robot - it's safe to say the failure rate is probably even higher than it is in industry. So, regardless of the researchers' intention, they just don't know if their design will be better.
In this case, "better" is defined as whether the new robot is able to perform in quantifiably superior ways in very specific environments. If it is better, it will become the standard on which future research is conducted; it may even find commercial use. If it isn't better, it will be forgotten.
This is equivalent to the survival of an organism in an ecology. Organisms evolve to fit their environment. If the environment changes too fast or too far, the organism won't survive. If a new type of organism evolves that performs better in that environment, it will eventually crowd the older type out of existence.
The thing is, though, the researchers don't actually know if their design will be successful (yet). They will test it to find out. Those tests are the equivalent of natural selection in biological organisms, and ultimately yield similar types of results: either the robot will become popular, or it will become extinct.
But even if it becomes extinct, it may leave "genetic" traces of itself behind, just like extinct lifeforms have left genetic traces of themselves in us. For the robot, those traces will be features, parts, and technologies that will recur in future robot designs - traces that, though they were part of a failed design, may themselves ultimately be beneficial in some future design.
Because of all this, then, the engineers and researchers who developed the robot are the source of mutation, the "reproductive system" of the robot, and a significant element of the "artificial selection" process (the socio-technical systems version of "natural selection") that leads to the robot's eventual success or extinction.
I hope this example helps show how we are part of the evolutionary process of our products.
1. http://www.jstor.org/stable/1250634?seq=1#page_scan_tab_contents
2. http://www.stage-gate.com/resources_stage-gate_latestresearch.php
3. http://hbswk.hbs.edu/item/clay-christensens-milkshake-marketing
4. https://newproductsuccess.org/white-papers/new-product-failure-rates-2013-jpim-30-pp-976-979/
No comments:
Post a Comment