The successful landing of NASA’s Curiosity on Mars is arguably the most significant robotics event of the decade. Not only have the stunning photographs at Bradbury Landing and the back story of the mission captured the imaginations of thousands of future scientists and engineers, but the value of robotics has been demonstrated to the public.
As a roboticist, there’s a lot to learn from the project. The most significant lesson is probably that you don’t have to wear a pocket protector to make a contribution to the field. NASA’s PR team has been highlighting the career paths of several key players in the project who don’t fit the expected stereotype roborocket scientist. For example, the leader of the Mars Lander component of the project is a would-be rock-and-roll star who barely passed high school. To satisfy his personal curiosity, he enrolled in a physics course at a community college, and his new trajectory was set.
Another lesson is the power of teams to tackle seemingly insurmountable obstacles. Although there were project leaders, the heavy lifting was performed by thousands of engineers and scientists over many years. Designing the Lander was a ten year project that at times involved 2,000 engineers, for example.
The mission also highlights the importance of simulation in the R&D process. The Lander was never subjected to a full physical test, but the design group relied on simulations and avoided a costly and lengthy build-crash-rebuild cycle.
Another take-away is start off by taking baby steps. After landing the six-wheeled vehicle on Mars, the scientists back at NASA didn’t take it for a remote joy ride across the Gale Crater. Instead, they tested the sensors (at least one wind sensor was found to be inoperative) and then made a short, slow, 16 minute excursion. There’s no room for error, and one wrong turn could result in a flipped vehicle and the end of the mission.
If I had to rank the lessons learned from the Curiosity project, I’d have to pick starting slowly and carefully as the most important takeaway. If you’re working with a $600 quadcopter and you accidently pilot it into a telephone pole at 40 mph, you’ll have to restart from scratch. Better to start off with small, short hops until you get a feel for the craft and the controller. It’s the same with a high-speed battlebot — you can’t rush testing when you’re dealing with high-speed spinning discs, hammers, and other tools of destruction.
From a strictly mechatronic perspective, there’s a wealth of information that you can gain by studying the images provided by NASA. For example, take a look at the propulsion system, including the wheels that can rotate freely about the vertical axis in a sort of ‘C Clamp’ cage. What have you borrowed from Curiosity’s design to make your next robot, whether it’s a simple terrestrial roamer, high-speed battle-bot, or a quadcopter a success? SV
Posted by Michael Kaudze on 11/08 at 02:11 PM