Project

# Title Team Members TA Documents Sponsor
19 Soccer Tracking Laser Gimbal
Aaron Zhang
Colin Smithsuvan
Jayden Kim
Wesley Pang
\#\#\# Soccer Tracking Laser Gimbal

Team Members:

- Colin Smithsuvan ([colin9@illinois.edu](mailto:colin9@illinois.edu))
- Jayden Kim ([jaydenk2@illinois.edu](mailto:jaydenk2@illinois.edu))
- Aaron Zhang ([aaronz4@illinois.edu](mailto:aaronz4@illinois.edu))

\# Problem
Typical sports broadcasting cameras are human controlled for tracking objects. We'd like to try to automate this with a vision-based system and track a soccer ball, specifically two people passing the ball to each other. This could help reduce human error and smooth out the motion of the camera and keep the object being tracked more centered in the filming camera's field of view. An extended scope of this project is tracking of an entire field.

\# Solution
We aim to solve this problem by having a camera positioned on a gimbal’s arm, which follows the movement of the soccer ball. This way, the ball is always positioned in the center of the camera frame, regardless of where the ball is being moved. For tracking the ball, we plan to deploy a vision detection model, which is trained to detect the soccer ball with real-time accuracy and latency. From there, we will have an arm with a camera at the base along with a laser which points at the ball for human verification. This arm will track the movement of the ball, which in turn will move the camera with it. This way, we can fully track where the ball is moving at all times with the ball in the center of the frame.

\# Solution Components
Laptop, Custom PCB, Motors, Power Supply

\#\# Subsystem 1
Laptop

- HP Pavilion Plus with RTX 4050 GPU and Intel Ultra 7 CPU with 1 TB storage and 32 GB of RAM.

\#\# Subsystem 2
Custom PCB

1) Power
1) Based on motor nominal voltage, the input bus voltage/dc link voltage will be 24V
2) To power the CAN transceiver, there needs to be 5V as characteristic of CAN
3) To support the logic level components like the microcontroller, gate driver logic supply, ethernet PHY, and others there will need to be a 3.3V rail
1) There will potentially be both an analog and digital 3.3V rail where the two are separated by a ferrite bead to keep any switching or digital noise from propagating onto the analog rail
4) To generate the 5V, 3V3\_D, and 3V3\_A rails, we plan to not design a custom switching converter circuit and buy a power conversion “brick”
2) Microcontroller
1) STM32H753
1) Contains CAN/CANFD controller, ethernet MAC, 16-bit ADCs, 16 and 32-bit high-resolution timers
2) Depending on number of ADCs and GPIOs needed can select from the different pin packages
1) ADCs and GPIOs will be used for sensing circuits, encoder interfaces, or telemetry we want
3) Inverter
1) Gate Driver
1) Smart gate driver IC \- DRV8304
2) H-bridge FETs
1) N-channel fets for the three H-bridges \- will need to size the FETS to fit within the maximum phase currents of the selected BLDCs (1.4A) \- 06N06L
1) 60V VDS, 5.5A
4) Communications
1) CAN
1) TCAN3403DRQ1
2) Ethernet
1) PHY \- LAN8742
2) RJ45 Connector \+ Magnetics

\#\# Subsystem 3
Motors

- The motors will be used to drive the motion of the two axis gimbal and allow the object to remain near center of the camera’s FOV
- Initially planned to use a brushless DC motor in order to learn the inverter stage design
- Fallback plan to use an integrated servo or brushed DC motor if unable to get brushless design working
- Motor will need an encoder interface for position tracking at low speed
- Motor PN: DFRobot 4015 3-phase brushless motor
- 24V, 1.4A

\#\# Subsystem 4
Power Supply

- Any power supply capable of 24V at \~10 A

\#\# Subsystem 5
Firmware

- State machine should look similar to this: INIT, CALIBRATE, IDLE, TRACKING, COAST (target lost, extrapolate briefly), SEARCH (scan pattern, FAULT)
- Input capture on camera trigger latches encoder counts at exposure where the timebase is shared with the laptop
- Potentiometer open-loop mode for inverter bring-up to test hardware before any vision/software code exists
- Telemetry streaming at full loop rate
- Drivers for Ethernet packing/unpacking, SPI, MAC, Auxiliary functions, timer, led controls, logic level gate control, etc

\#\# Subsystem 6
Vision Model

- Any on-device model for real-time streaming that can accurately provide bounding boxes. We will take these bounding boxes and take the middle of the bounding box for the location that the laser should point to. The SOTA model that we will try first is YOLOv26 since it provides fast on-device inference and is primarily trained on datasets such as COCO, which include a class for “sports ball.”
- We may also need the camera intrinstics and extrinstics to map the image pixel coordinates to real world coordinates to understand how much to move the turret arm in the real world relative to the image detected bounding box.
- In order to map from pixel coordinates to real world coordinates, our biggest challenge is accurate depth estimation which is required for perspective projection. Single camera depth estimation is an on-going research challenge with many models not being industry ready. Therefore, we may have to implement two gimbals for triangulation or to use a simple lidar for more accurate depth estimation for backup. However, our primary method of approach is to try monocular depth estimation models for our single camera use case.

\# Criterion For Success
Primary Objective: Demonstrate a closed loop, vision guided camera platform that keeps a moving target centered in frame.

High Level Goals

- Standalone vision model provides accurate bounding boxes on a soccer ball on laptop
- Camera can successfully send gimbal position data to the MCU
- MCU successfully ingests data and passes to the laptop
- Laptop can run vision model inference and send position of where the gimbal should point back to the gimbal
- Gimbal is able to move within a 120 degree field of vision both vertically and horizontally to match the position of the object detected
- Keep a moving target centered in frame at all times autonomously
- Gimbal is able to move fast enough to keep the target in frame.
- Inference is developed with low-enough latency to have a fast closed-loop

Oxygen Delivery Robot

Aidan Dunican, Nazar Kalyniouk, Rutvik Sayankar

Oxygen Delivery Robot

Featured Project

# Oxygen Delivery Robot

Team Members:

- Rutvik Sayankar (rutviks2)

- Aidan Dunican (dunican2)

- Nazar Kalyniouk (nazark2)

# Problem

Children's interstitial and diffuse lung disease (ChILD) is a collection of diseases or disorders. These diseases cause a thickening of the interstitium (the tissue that extends throughout the lungs) due to scarring, inflammation, or fluid buildup. This eventually affects a patient’s ability to breathe and distribute enough oxygen to the blood.

Numerous children experience the impact of this situation, requiring supplemental oxygen for their daily activities. It hampers the mobility and freedom of young infants, diminishing their growth and confidence. Moreover, parents face an increased burden, not only caring for their child but also having to be directly involved in managing the oxygen tank as their child moves around.

# Solution

Given the absence of relevant solutions in the current market, our project aims to ease the challenges faced by parents and provide the freedom for young children to explore their surroundings. As a proof of concept for an affordable solution, we propose a three-wheeled omnidirectional mobile robot capable of supporting filled oxygen tanks in the size range of M-2 to M-9, weighing 1 - 6kg (2.2 - 13.2 lbs) respectively (when full). Due to time constraints in the class and the objective to demonstrate the feasibility of a low-cost device, we plan to construct a robot at a ~50% scale of the proposed solution. Consequently, our robot will handle simulated weights/tanks with weights ranging from 0.5 - 3 kg (1.1 - 6.6 lbs).

The robot will have a three-wheeled omni-wheel drive train, incorporating two localization subsystems to ensure redundancy and enhance child safety. The first subsystem focuses on the drivetrain and chassis of the robot, while the second subsystem utilizes ultra-wideband (UWB) transceivers for triangulating the child's location relative to the robot in indoor environments. As for the final subsystem, we intend to use a camera connected to a Raspberry Pi and leverage OpenCV to improve directional accuracy in tracking the child.

As part of the design, we intend to create a PCB in the form of a Raspberry Pi hat, facilitating convenient access to information generated by our computer vision system. The PCB will incorporate essential components for motor control, with an STM microcontroller serving as the project's central processing unit. This microcontroller will manage the drivetrain, analyze UWB localization data, and execute corresponding actions based on the information obtained.

# Solution Components

## Subsystem 1: Drivetrain and Chassis

This subsystem encompasses the drive train for the 3 omni-wheel robot, featuring the use of 3 H-Bridges (L298N - each IC has two H-bridges therefore we plan to incorporate all the hardware such that we may switch to a 4 omni-wheel based drive train if need be) and 3 AndyMark 245 RPM 12V Gearmotors equipped with 2 Channel Encoders. The microcontroller will control the H-bridges. The 3 omni-wheel drive system facilitates zero-degree turning, simplifying the robot's design and reducing costs by minimizing the number of wheels. An omni-wheel is characterized by outer rollers that spin freely about axes in the plane of the wheel, enabling sideways sliding while the wheel propels forward or backward without slip. Alongside the drivetrain, the chassis will incorporate 3 HC-SR04 Ultrasonic sensors (or three bumper-style limit switches - like a Roomba), providing a redundant system to detect potential obstacles in the robot's path.

## Subsystem 2: UWB Localization

This subsystem suggests implementing a module based on the DW1000 Ultra-Wideband (UWB) transceiver IC, similar to the technology found in Apple AirTags. We opt for UWB over Bluetooth due to its significantly superior accuracy, attributed to UWB's precise distance-based approach using time-of-flight (ToF) rather than meer signal strength as in Bluetooth.

This project will require three transceiver ICs, with two acting as "anchors" fixed on the robot. The distance to the third transceiver (referred to as the "tag") will always be calculated relative to the anchors. With the transceivers we are currently considering, at full transmit power, they have to be at least 18" apart to report the range. At minimum power, they work when they are at least 10 inches. For the "tag," we plan to create a compact PCB containing the transceiver, a small coin battery, and other essential components to ensure proper transceiver operation. This device can be attached to a child's shirt using Velcro.

## Subsystem 3: Computer Vision

This subsystem involves using the OpenCV library on a Raspberry Pi equipped with a camera. By employing pre-trained models, we aim to enhance the reliability and directional accuracy of tracking a young child. The plan is to perform all camera-related processing on the Raspberry Pi and subsequently translate the information into a directional command for the robot if necessary. Given that most common STM chips feature I2C buses, we plan to communicate between the Raspberry Pi and our microcontroller through this bus.

## Division of Work:

Given that we already have a 3 omni wheel robot, it is a little bit smaller than our 50% scale but it allows us to immediately begin work on UWB localization and computer vision until a new iteration can be made. Simultaneously, we'll reconfigure the drive train to ensure compatibility with the additional systems we plan to implement, and the ability to move the desired weight. To streamline the process, we'll allocate specific tasks to individual group members – one focusing on UWB, another on Computer Vision, and the third on the drivetrain. This division of work will allow parallel progress on the different aspects of the project.

# Criterion For Success

Omni-wheel drivetrain that can drive in a specified direction.

Close-range object detection system working (can detect objects inside the path of travel).

UWB Localization down to an accuracy of < 1m.

## Current considerations

We are currently in discussion with Greg at the machine shop about switching to a four-wheeled omni-wheel drivetrain due to the increased weight capacity and integrity of the chassis. To address the safety concerns of this particular project, we are planning to implement the following safety measures:

- Limit robot max speed to <5 MPH

- Using Empty Tanks/ simulated weights. At NO point ever will we be working with compressed oxygen. Our goal is just to prove that we can build a robot that can follow a small human.

- We are planning to work extensively to design the base of the robot to be bottom-heavy & wide to prevent the tipping hazard.