Page 38 - Handbook of Biomechatronics

32 Reva E. Johnson and Jonathon W. Sensinger

Fig. 1 Block diagram of typical open-loop (A) and closed-loop (B) control system. The
actuator receives a control signal that dictates how the supplied energy should be
converted into a mechanical movement or force that acts on the plant or process.

2 DESIGN GOALS OF ACTUATORS

Below we discuss three broad design goals that apply to every
biomechatronic actuator: safety, performance, and ease of use. Within each
broad design goal are specific metrics, whose desired values depend on the
purpose of the overall system. These metrics help quantify the trade-offs of
design choices. For example, there is often a trade-off between safety and
performance. One strategy to improve actuator safety is to decrease the stiff-
ness, so that interaction with humans is more flexible and injury-causing
impacts are minimized. However, a decrease in stiffness can also worsen per-
formance by reducing bandwidth. When faced with this common trade-off,
how can we minimize injury while still designing a useful actuator?
Quantitative metrics enable us to optimize the system for several design
goals. One example of a design optimization for a PUMA 560 robot is
shown in Fig. 2. The PUMA 560 is an articulated robot, originally designed
for industrial assembly lines and now widely used for research and education.
The PUMA often operates alongside or directly interacts with humans; so,
minimizing injury risk is an important design goal. Fig. 2B is an example of
how a plot can be used to show how design parameters (in this case, actuator
stiffness and effective inertia) influence design metrics (in this case, head
injury risk). The designer then selects a combination of parameters to
achieve the desired outcome metric. Similar multivariable optimizations
can be used to choose the type and characteristics of other biomechatronic
actuators (another example is provided in Fig. 3B).

2.1 Safety

How do I design an actuator to interact with humans safely?