Mars Science Laboratory is a mission to create and send to Mars the largest, most capable and most exciting rover that has been sent to another planet to date. It will be a remote robotic scientist that will help us investigate our most Earth-like neighbor in the solar system. It is literally a mobile laboratory — the size of a car, with a wide array of science instruments that will help us determine whether Mars has the capability to support life, both in the past and in the present.
Most of my work at JPL has been in the area of research robotics, small projects with very focused goals, such as the Urban Robot project, the Spiderbot, and many others. These robots were created using a small team of engineers who each covered a wide area of responsibility such as mechanical, electrical, and different areas of computer science for perception and navigation.
At times, to get one of these robots working right we would “hack” together a solution, and get it implemented in a very short amount of time! By throwing our energy and ideas into each project, we could push the cutting edge of different technologies and robotic capability that could then be used by future projects and researchers.
In contrast I’m now working on the motor control system of the Mars Science Laboratory mission, which as opposed to research is a “flight project” (it’s going to fly!). This is quite a different experience than the research area - the mission is kind of like taking the goal or purpose of the robot, breaking it down into a million pieces, and putting every piece under a microscope to make sure everything will work absolutely perfectly. Instead of 10 or fewer engineers each working on the different subsystems of the robot, we have hundreds of engineers and scientists who are planning, designing, developing, manufacturing, testing and in general creating a very complicated remote-sensing system.
Sending a robot of any kind to another planet is a completely different story than running any such thing on Earth. For one thing, the robot must operate within very extreme temperatures and handle harsh exposure to the sun’s rays. Future robots may have to deal with steep or challenging terrain, or even a lack of a solid surface such as on Titan, Saturn’s largest moon. And throughout all this, the robot has to work perfectly and stay in communication with Earth. There is no control-alt-delete button to handle software crashes, and no technician around who can run out to push the big red reset button. In other words, if you’re going to run a robot on another planet, it has to land unharmed, work the first time, and run correctly every time you command it to do something, so that you’re guaranteed to get back the vital science data you’re after.