The head acts as a natural focal point for nearby workers. The lighting interface gives nearby workers an immediate visual indication of Atlas's operational state, intended actions, and any system warnings.
Why Boston Dynamics Designed Atlas' Head This Way
Q&A with Taylor Frey-Baker, Senior Mechanical Engineer at | Boston Dynamics
Tell us about yourself and your role at Boston Dynamics working on the Atlas program.
I’m a mechanical engineer on the Head team for Atlas. My primary focus is a subassembly inside of the head, the compute core. Essentially, I work on the brain of the robot.
What drove the decision to give Atlas a distinctly non-human head design rather than a more human-like face?
Our design choices are driven by function, performance, and efficiency rather than trying to replicate human aesthetics. Atlas will be a robot working in industrial environments, so it should look like an industrial piece of equipment rather than a human being. Even though Atlas’ head isn’t meant to resemble a human head, it does have features which communicate intent in an intuitive way. The round light rings at the front and back of the head can indicate the robot’s status as it works, and can be seen from across a factory floor. Atlas’ head can rotate and tilt so that the front face indicates where the robot is focusing.
How do the head-mounted HDR stereo cameras improve situational awareness, particularly during teleoperation in dynamic environments?
The original hydraulic Atlas was primarily a research platform focused on athletic maneuvers, balance, and full body mobility so it only needed to see the ground directly in front of it to judge terrain. Our electric Atlas is designed for real-world commercial applications alongside people, which demands much broader situational awareness. Having the cameras in the head gives Atlas a significantly wider field of view.
The high definition cameras in the head give remote operators a higher-resolution view of Atlas’s surroundings, and the HDR (high dynamic range) capabilities allow an operator to see in a wider range of lighting conditions. For example, the HDR camera makes it easier to see the inside of a dark storage compartment that is backlit against a bright window (the image doesn’t become washed out), which in turn makes it easier to retrieve an object from inside the compartment.
How do the head's status and intent lights communicate the robot's planned actions or state to nearby human coworkers?
Because Atlas is built to operate in shared industrial spaces, clear communication is essential for workplace safety. The head acts as a natural focal point for nearby workers. The lighting interface gives nearby workers an immediate visual indication of Atlas’s operational state, intended actions, and any system warnings. Because the head is a natural focal point and can rotate continuously, workers can also quickly see where Atlas’s sensors are focused and understand where its attention is directed, helping reduce ambiguity in shared industrial environments. For instance, the front and rear light rings can illuminate in different colors or blink in patterns to indicate the robot’s status.
How does this overall head design facilitate safe and efficient collaboration on factory and warehouse floors?
The head design acts as a two-way bridge for safety and efficiency. On the one hand, the elevated stereo perception system gives Atlas the spatial awareness it needs to navigate unexpected obstacles and perform complex tasks in a safe and reliable way. On the other hand, Atlas can also continuously communicate its state to nearby workers using lights, audio feedback, and head orientation. Together, these features help Atlas fulfill our mission of taking on repetitive, strenuous, or dangerous tasks as well as work safely, predictably, and efficiently alongside people in shared factory and warehouse environments.
Taylor Frey-Baker is a senior mechanical engineer at Boston Dynamics, where she contributes to mechanical design and thermal management efforts for the Head team for Atlas. She has previously worked on the consumer product design of robotic vacuum cleaners and animatronic toys, and she has experience in engineering consulting across the medical, commercial, and defense industries. She has an M.S. in Product Design & Manufacture from Boston University, and a B.S. in Mechanical Engineering from Northeastern University.
The content & opinions in this article are the author’s and do not necessarily represent the views of RoboticsTomorrow
Featured Product
