Five Robotics Failures and What Developers Can Learn From Them

Progress is not inevitable. Instead, it is the product of hard work, careful decisions, and applying lessons from failures. Rather than indulge in schadenfreude over recent robotics setbacks, let's see what developers, integrators, and users can learn from them.

It's easy to assume that robots, drones, and autonomous vehicles are becoming ever more capable. Certainly, sensors have become cheaper, artificial intelligence algorithms more sophisticated, and automation more widespread across industries. We've shared plenty of news stories about how automation is enabling the current e-commerce boom and how robotic and drone delivery trials are progressing.

At the same time, smart speakers overtook several social robots, most agricultural systems are still in the prototype phase, and today's videos of dancing humanoids only came after years of failures such as those in the DARPA Robotics Challenge.

Over the past year, many companies laid off staffers, but relatively few shut down. Here are some recent failures and possible lessons to be drawn from them.