The following image depicts a high-level system diagram of the electrical system. This drawing includes the major functions of the custom printed circuit board (PCB). The items marked with a purple flower are nice-to-have features. Conversely, the bee indicates a critical feature, required for proper functionality.

MOT-RB37GM-150 brushed DC motors were chosen to move the robot’s entire frame. These motors can handle 2 Nm at 30 rpms.
The motors were selected based on a series of operating condition statics and dynamics analyses. The following load conditions were used:
The second step included force analysis of the load cases through free body diagrams. After the free body diagrams, Newton’s second law was applied to calculate the forces and torques of the system.

A half-bridge motor controller (DRV8906QPWPRQ1) was chosen to drive the DC brushed motors and stepper motors. Each chip contains 12 half bridges; therefore, it can drive 6 motors. Two chips will be used: one chip will drive the four DC brushed motors and the other chip will drive the three stepper motors. The chips will communicate to the central processor via a SPI bus. This allows for the chip to receive speed commands and send error messages. The chip can handle up to 35 V, but it will only be supplied 24 V. Additionally, the chip provides over current protection features (OCP). This is ideal to manage motors that have very high stall currents. When the system motors stall, it is estimated that they will draw about 12 A. Each half bridge can only handle 1.8 A before the OCP fault occurs. This means that the chip will shut down and enter a faulted state at a current far lower than the total stall current. This will protect the chips and the system from the 12 A stall current. As explained in the power tree section, a 4.5 A fuse will be used to protect the 24 V rail.

The most important sensor is the Intel RealSense D405 Camera. This camera will act as the eyes of the robot to help it locate the centers of the flowers. The camera will also aid in orienting and moving the brush to the center of the flowers. This camera was chosen for its compact casing, easy integration with an NVIDIA Jetson Nano processor, relatively low cost, RBG and depth data capabilities, as well as its high accuracy over the desired range. The camera will connect to the Jetson Nano with a micro-B-USB to USB-A cable.
The four encoders (AMT102-V) will be used to detect wheel speed, approximate robot turning radii and approximate distance travelled. These encoders have a mating set up that allows them to be mated around the motor shaft. The motor shaft will turn the encoder’s mating bearing to obtain encoder values. The output quadrature signals will be processed by the microcontroller unit.
Encoders are prone to inaccuracy when the wheels slip. It is expected that the wheels might slip on loose soil or muddy terrain. An IMU (MPU-9250) will be used to aid in locating the robot’s rotation.
The thermistor (RL0503-5820-97-MS) and a voltage sense circuit will facilitate the battery management system (BMS). The thermistor will sense the temperature that the battery is running at. The temperature and voltage readings will allow for the following battery safety features: