The stability of a 4 bar linkage made it easy to use a single servo to lift the CRATE. It allows for a controlled motion, mechanical advantage, stability, and was efficient with only needing one servo. It seems easily enough to provide one command to the servo to bring the CRATE to the desired height and was easily repeatable. The thick acrylic (1/4") helped in stabilizing the lifting mechanism and also made it space efficient to allow for it to stretch up high when placing CRATEs and staying in tight at the beginning of a match or when not in use. With three ME students on the team, this also seemed like the best option.
The 4 bar linkage then had the "crate manager" up front which lifted the CRATEs using a press fit magnet and a 9g servo that pushed out the CRATE to place on the STACK
The primary functionality of the left and right towers were to hold the lifting servo and the IR base that held our signal conditioning circuit. These towers also held our buck converters and power switch
The towers were held in place by the tower clamps. These clamps clamped the boards onto the chassis top piece.
The chassis top had holes for standoffs to hold the bulk of the boards: PIC32 boards, power distribution boards, motor driver boards. This piece had an "open" design to allow for various wires. This also held the TOF mount that allowed for our bot to measure the distance from walls.
The 4 spacers were placed in a rectangular configuration and made with a small radius to minimize their obstruction. Heat set inserts at the top and bottom were used to wedge these spacers between the top and bottom chassis.
The chassis bottom piece held the drivetrain, batteries, servo flags, and IR sensors for tape sensing. The shape of this piece was curved towards to front to prevent from getting stuck in corners and slighly decrease weight.
The battery mounts were designed to keep the batteries from sliding around but not so much that they were difficult to remove to recharge.
The 5 IR sensors at the front were held in place with the IR sensor mount that was screwed into place. These 5 IR sensors were placed as far forward as possible to allow for easier line following.
The wheels chosen were 72 mm, 30A durometer wheels to allow for a high amount of traction and thereby decrease any slipping. These wheels were great when clean, but the field constantly made them dirty which sometimes caused them to slip. This was only an issue while testing and debugging as we had our wheels cleaned for checkoffs, grading, and the competition.
The pillow blocks held small bearings that minimized the footprint of the drivetrain and reduced the radial load on the motor shaft.
5mm flange couplings connected the motor shaft to some flange adapters that connected the flange couplings to the wheels on both sides.
The motors mounts had screw holes to hold the motor in place and onto the chassis.
The servo flags on the right and left had blue and green flags controlled by 9 g servos that denoted the game start, side, and end.
With only two motors to drive the bot, we decided to put the wheel as far back as possible from the IR (tape) sensors and use pillow blocks to prevent the radial loading on the motor shaft.
At the beginning, the lift is kept in tight to stay within the starting bounding box of 32x32x32cm.
When barking, Frank lifts up a CRATE to be in line with the highest compartment in the STACK.
When barking a little less lound, Frank lifts up a CRATE to be in line with the middle compartment in the STACK.
The three linkages shown above are made of laser cut acrylic and were optimized to allow for the maximum allowable height while keeping Frank within the required dimensions. Their width was chosen based on the servo attachment and desired height. The fourth linkage is completed by the wooden dowl. The linkage's lateral movement is constrained by shaft collars. These collars also stiffen and provide consistent movement of the linkages. At the front, the shaft collars also held the crate manager in place. Bearings pressed into the linkages allowed for the 4 bar to move freely and with as little friction as possible to allow the servo to provide the desired position.
Laser cut. 1/4" acrylic. This laser cut chassis held in place by 4 spacers. This piece also has holes to mount the PIC32 boards, power distribution board, and motor driver boards. The front of this piece also holds the left and right towers that have a tight fit into this piece.
Laser cut. 1/4" acrylic. This other laser cut chassis piece holds the drivetrain, battery mounts, servo flags, and IR sensor.
Laser cut. Duron for flexibility. Connects the flange coupling to the wheel.
Laser cut. 1/4" acrylic. Holds the signal conditioning circuit to read the beacon.
Laser cut. 1/4" acrylic. Holds the left side of the IR base, a buck converter and has holes for wire management.
Laser cut. 1/4" acrylic. Holds the right side of the IR base, a buck converter, the game switch, has holes for wire management and holds the lift servo 3d printed bracket.
Laser cut pieces. 1/4" acrylic (for stiffness) pieces with press fit bearings to allow for the 4 bar linkage to lift the crate manager.
3D printed. Holds the motors securely using the motor screw holes and mounts to the chassis.
3D printed. Holds our 5 IR sensors up front for more efficient line following.
3D printed. Provides space for the batteries and motor mounts while allowing for easier cable management.
3D printed. Take most of the radial loading away from the motor shaft.
3D printed. Firmly hold batteries in place while allowing for easy removal and access.
3D printed. Has a designated space big enough for a small magnet to pick up the CRATE, mounting holes for the pusher servo to push CRATE off, holes to attach to the 4 bar linkage, and stabilizing pins to prevent the CRATE from rotating and to pick up in desired CRATE orientation. The pusher is a 3D printed part that attaches to the 9g servo.
3D printed. Determine game start, game end, and game side. Blue and green tape was added to corresponding flag.
Clamps to keep left and right tower in place. Screws from the top make it easier to remove.
3D printed. Held TOF sensor at the front of the bot to measure the distance from walls to determine when to deposit cube.
3614 Series Rhino Wheel (14mm Bore, 72mm Diameter, 30A Durometer) from gobilda.
Drive our motors. More in electrical page!
Brains of the operation. More in electrical page!
Provide power to the servos. More in electrical page!
Provide a constant voltage input to our PDB. More in electrical page!
Provide power to the circuit. More in electrical page!
Provide a clean, readible output from the IR sensor input. More in electrical page!
5 sensors used for tape sensing, used as analog inputs. By placing these 5 in line at the front we were able to determine a gradient for the line following algorithm.
2 servos used to actuate flags and 1 used to push cube off of the crate manager.
Provided by lab. Used to move our wheels and our entire bot.
note: image is not actual motor, but a close representation (had a nice white background)
Powerful servo used to move our 4 bar linkage.
TOF sensor was chosen over ultrasonic due to it having greater accuracy, more range, and slightly faster readings. It was also great not having to deal with needing to be parallel to a wall to obtain nice readings as is needed for ultrasonics.
Click the link to download the full assembly!