Saturday, December 17, 2011

In pursuit of Ackerman-ish steering






I am in the middle of laying out the steering linkage for the motorcycle-engined remote control car. I decided that I wanted to try and use a low-cost, high force, linear actuator to control the steering. I can find some for a little over $100-$200 dollar range that can be controlled using serial commands. My hope is this type of actuator would be more robust to mechanical shock than a conventional RC servo. Industrial RC servos exist, but they are twice the cost for anything that I think has the torque to steer my car over rough terrain. I am also skepital of their robustness to mechanical shock. For the first iteration of the car I plan to implement Ackerman steering. The linkage typically used for approximating Ackerman steering is shown here:


I also decided, for aesthetic reasons, I would like to place the axis of the linear actuator along the longitudinal axis of the car. So now I am trying to see if this can be done in a reasonable manner. Basically I need to transmit force originating along the longitudinal axis of the car into right left motion of the coupler link of the Ackerman steering linkage. I did some thinking and I decided to make some popsicle stick models to try and spark a breakthrough. I broke out the cordless drill and got to work.

At this point, I have not solved this challenge just yet, but I did come up with some interesting things in the process. First off, a couple years ago I came up with this single DOF linkage. It was the inspiration for my new popsicle stick models.


The unique thing about this linkage is that a small angular motion at the top of the linkage results in a rapid deployment of the rest of the linkage along the medial axis. As I was trying new linkage ideas today I designed this guy.




This is a 14 bar, single degree of freedom linkage. I think it looks kind of neat. One interesting thing I discovered about it, is that it has two of the sub four-bar linkage ages that make up the composite linkage have reachable toggle positions. The picture below shows the links I am referring to.



An interesting observation I made, is that since it is a single degree-of-freedom linkage, it is possible to block the movement of the linkage simply by placing one of the two sub-system four bar linkages into a toggle position. This could have interesting applications in larger linkage systems with multiple degrees of freedom. You could essential inactivate degrees-of-freedom of the composite linkage by incorporating sub linkages that feature reachable toggle positions. You could also in some sense activate new states of the linkage by forcing the linkage past its toggle positions. I am not sure you see many complicated linkage systems around these days but I thought it was interesting. It could make some interesting artwork.