Ryan Bao - Biorobotics

Biorobotics lab - VErT project

The VeRT Project focuses on developing a prototype to serve as a baseline in an upcoming DARPA Grand Challenge. It focuses on building a robot that can maneuver complex terrain such as stairs or rocky terrain. The goal of the project is to build the robot such that it is able to climb stairs faster than a firefighter while carrying 100 pounds, the average weight of a firefighter's gear.

U0 Prototype

The U0 Prototype was designed to be a baseline characterization robot. It was designed to have interchangeable wheel and tread designs, so that we could characterize various designs and determine the optimal one. There are also several innovative designs on the prototype to improve performance

Movable payload to adjust center of gravity
High ground clearance
Adjustable treading to increase or decrease contact area with the ground depending on activity
Terrain detection using cameras and LiDaR

PXL_20221220_204811552.mp4

Machining support

A project of this scale requires significant hands-on support. I performed the following tasks.

Soldering
Tapping
CNC Milling
Manual Milling
Lathing

Sensor Payload Mounting

I was responsible for designing an intentionally cantilevered support for a sensor payload consisting of a LiDaR and 8 cameras. This payload mounting also had to be 15 inches above the movable payload to prevent interference. In addition, the only mounting points that were available to me were on the front plate of the robot.

The payload was relatively light. Because of this, I decided to build the structural components out of 80/20 extruded aluminum, which was readily available to me and simple to mount to. In order to ensure that the 80/20 was strong enough to support the payload, I used basic hand stress analysis and confirmed with FEA.

The front plate was made of 1/4" thickness aluminum plate, cut with a CNC router.

Physical Payload Mounting

I helped design a payload that was movable parallel to the drive train, so that we could modify the center of mass of the overall robot to increase stability as it traversed difficult terrain. Our criteria for the design was

Needed to be able to move quickly enough to increase stability of the robot when moving at low-medium speeds (5-10 mph)
Needed to be able to support the entire payload weight (100 lbs)

Our first design involved using two linear rails and a lead screw to move the payload back and forth. We encountered two critical faults with this design. The first, and most critical, was that the payload did not move fast enough. Generally speaking, lead screws are not fast enough for this sort of application. The second is that it was difficult to keep the rails aligned on 80/20 aluminum.

To fix the speed issue, we switched to a linkage mechanism. We haven't found a convenient fix for the alignment issue, although we are looking into it.

Page updated

Google Sites

Report abuse