Design Ideas and Prototypes

Prototype 1

For the first design, complex linkage systems that combine basic linkage systems (like two 4-bar linkage or a 4-bar-slider-crank linkage) were considered, instead of trying to further improve the 4-bar linkage. The reason was that in order to fulfill the design specifications, the trajectory has to be slightly more complex than the regular ones from the 4-bar linkages.

The first prototype was then created as a slider crank mechanism with a slot that constrains the motion of the coupler link to match the intended trajectory (ideally an 'S' shape tracing the sides of the robot).

The CAD design showed promising signs of matching the trajectory; however, practical challenges such as the manufacturing of the link and the forces of the pin on the slot were considered, and was replaced by a different design.

The CAD model of the first prototype. It produced a trajectory that was close to the ideal trajectory, but there were a lot of concerns in the manufacturing and the smoothness of the slot movement.

Prototype 2

The next prototype was then aimed to achieve greater extension from the robot on the lateral direction as possible and a minimal movement parallel to the robot – as to reduce the complexity of the mapped area problem. Several mechanisms with this behavior were researched upon that led to a wing retraction mechanism, like bird wings. Birds that are not in flight retract their wings close to the body while they expand their wings during flight.

Based on that concept, the final design was then created as a slider crank mechanism with a four bar linkage mounted to one of the links of the slider. Since a one DOF motion is desired, we define the input angle of the four bar as the complementary angle of the slider. This feature was desirable since it could reach up to 6 inches away from the robot., which could amount to a mapped surface of about 128 in² (while the robot is stationary), as well as < 1 in of parallel travel area.

Initially the motor was assigned to drive link E to determine the angle φ. Links E, F and G would move the ground link d from the four bar mechanism. The angle between link E and G would define the input angle α that would allow  for the four bar mechanism. After a few simulations, we realized the accuracy of the input angle φ has to be high. Hence the input was set to be the distance of link F (i.e. the slider becomes the driving link), since the stepper motor would not be able to achieve that resolution.

The lengths of the links were tweaked to maximize the lateral distance (in the x-direction) and to minimize the parallel distance (in the y-direction). These figures below are some examples of the combinations we analyzed:

The dotted lines represent the trajectory of the sensor (cyan circles).

The links were then designed to be aluminum rods with steel rod ends attached. The slider was designed to move via a threaded Acme screw driven by the motor. The joints of the links were decided to be pins initially, but shoulder bolts were used instead because of the affordability over the pins.

The CAD model drawn in Solidworks. The figure on the left shows the mechanism begin attached to the base of the robot.