CAD's Role in Engineering Next Gen Robotics

Note: this is final post in a two-part interview series with robotics engineer and visionary Donald Hutson. Read part one to learn how programs such as BattleBots and FIRST Robotics are introducing students to the exciting side of engineering.

If you feel like Donald Hutson has been in your living room before, he has. Donald’s Mutant Robots team designed and piloted both the reining two-time Super Heavyweight champion “Diesector and the newly created Lock-Jaw Robot. Donald feels that SOLIDWORKS is a valuable weapon to have in the complex mechanical design world that is BattleBots. “SOLIDWORKS is going to thrive in this type of mechanical competition,” Donald said. “It helps me build the most complex machine I possibly can.”

While his team fell short of a third championship in the 2015 robotic combat tournament, chalk it up to using a cast bracket instead of a machined piece. Donald and Mutant Robots are already designing a new set of jaws with shock absorption and custom sprockets to solve the problem. This update will include running an FEA to make sure the same problem doesn’t fail Lock-Jaw again. The champ is on the comeback trail – hey, even Rocky lost a few fights.

Lock-Jaw is just one robot Donald designed in SOLIDWORKS. He actually completes all of his personal and professional projects in the software. Donald starting using SOLIDWORKS in 2000 and now his daily work always begins by firing it up. “I used to think it wasn’t possible to use a design tool to think through the process, but now I can’t do without it,” Donald explained. “It helps me quickly iterate on designs, thinking ‘that looks cool, how about his other concept?’”

Amazingly, Donald regularly lays out five or six different concepts in SOLIDWORKS at a time to evaluate the positives and negatives in very complex assemblies. “This helps me see if I’m headed down the right road and interface with other departments around my projects,” Donald said. The ability to intertwine different disciplines into projects has been a real enabler for Donald. In the past, engineers, drafters, and machinist were all very different people. Donald has experienced a change in the last decade. “Now SOLIDWORKS and CAMworks are providing enabling technology, allowing designers to also be machinist,” Donald explained. “The result is better engineers and designers because it gets them to think about how other elements fit and work together.”

At Qualcomm, Donald works as a robotics engineer, where he’s designed the Snapdragon Cargo, a flying and driving drone that picks up objects and can drive over rough terrain. complete with an integrated Snapdragon flight controller. The Snapdragon does some amazing maneuvers – including driving up and down stairs and even driving on ceilings. Why be a fly on the ceiling when you can be a Snapdragon? Previously he worked at the Neurosciences Institute where he focused on brain-based robotic platforms. Here, he was most interested in building robots that simulate the inner workings of a brain to better understand how our minds work.

To date, Donald’s work has been approaching two key elements that will drive the next generation of robotics: perception and communication. “These factors are part of a big push to add more sophistication to make drones safer and smarter,” Donald said. Essentially, next-gen drones will need to visualize and understand their environments, then effective relay this information to a pilot in order to have a human in the loop if the robot has a level of autonomy. As you can see in the below video, the Snapdragon Cargo is a great example of this approach.

“Machine vision and things like 3D point clouds are hard problems to solve,” Donald stated. “Drones need to take in a scene and put some context into the situation. The drone isn’t seeing a yellow block; it’s seeing a fire hydrant.” Without a proper understanding of its surrounding, a drone cannot begin to move and make decisions quickly. If drones are to become more complex and versatile, perception and communication must continue to be high priorities for innovation. Donald puts it best, “you can’t make a good decision if you can’t see, sense, and compute before you act.”