Robotics Commons^™

Robot Navigation In Cluttered Environments With Deep Reinforcement Learning, Ryan Weideman

Master's Theses

The application of robotics in cluttered and dynamic environments provides a wealth of challenges. This thesis proposes a deep reinforcement learning based system that determines collision free navigation robot velocities directly from a sequence of depth images and a desired direction of travel. The system is designed such that a real robot could be placed in an unmapped, cluttered environment and be able to navigate in a desired direction with no prior knowledge. Deep Q-learning, coupled with the innovations of double Q-learning and dueling Q-networks, is applied. Two modifications of this architecture are presented to incorporate direction heading information that …

Comparison Of Lqr And Lqr-Mrac For Linear Tractor-Trailer Model, Kevin Richard Gasik

Master's Theses

The United States trucking industry is immense. Employing over three million drivers and traveling to every city in the country. Semi-Trucks travel millions of miles each week and encompass roads that civilians travel on. These vehicles should be safe and allow efficient travel for all. Autonomous vehicles have been discussed in controls for many decades. Now fleets of autonomous vehicles are beginning their integration into society. The ability to create an autonomous system requires domain and system specific knowledge. Approaches to implement a fully autonomous vehicle have been developed using different techniques in control systems such as Kalman Filters, Neural …

Localization Using Convolutional Neural Networks, Shannon D. Fong

Computer Engineering

With the increased accessibility to powerful GPUs, ability to develop machine learning algorithms has increased significantly. Coupled with open source deep learning frameworks, average users are now able to experiment with convolutional neural networks (CNNs) to solve novel problems. This project sought to train a CNN capable of classifying between various locations within a building. A single continuous video was taken while standing at each desired location so that every class in the neural network was represented by a single video. Each location was given a number to be used for classification and the video was subsequently titled locX. These …

Baseball Shagger, Anthony Velasquez, Nick Walker, Nathaniel Kai Paresa

Computer Engineering

The purpose of our project is to allow players to hit baseballs on a baseball field and not have to worry about picking them up. By combining our knowledge of software and hardware, we developed the first design of a robot that “shags” baseballs. Our endeavor was only partially successful. The device was tested on grass, turf, and concrete. The motors did not have enough torque to get moving on grass. The device faired better on turf where it could move, but was quite jerky as the motor drive needed to be high to start moving, but once it was …

Towards Autonomous Localization Of An Underwater Drone, Nathan Sfard

Master's Theses

Autonomous vehicle navigation is a complex and challenging task. Land and aerial vehicles often use highly accurate GPS sensors to localize themselves in their environments. These sensors are ineffective in underwater environments due to signal attenuation. Autonomous underwater vehicles utilize one or more of the following approaches for successful localization and navigation: inertial/dead-reckoning, acoustic signals, and geophysical data. This thesis examines autonomous localization in a simulated environment for an OpenROV Underwater Drone using a Kalman Filter. This filter performs state estimation for a dead reckoning system exhibiting an additive error in location measurements. We evaluate the accuracy of this Kalman …

Senior Project - Roborodentia Robot, Nicholas Alexander Ilog

Computer Engineering

This project includes an autonomous robot capable of dispensing balls from a dispenser mounted on a wall and shooting the balls through targets five to eight feet away. The robot can hold up to five balls at a time and shoots balls one by one at targets.

Roborodentia Robot: Treadbot, Stephen C. Schmidt

Computer Science and Software Engineering

This document is a summary of my contest entry to the 2018 Cal Poly Roborodentia competition. It is meant to be a process overview and design outline of the mechanical, electrical, and software components of my robot.

Dynamic Model For Simulating Motion Of The Right Ventricle, Brian Michael Larsen, Sam Koopman Porter, John Francis D'Ambrosio

Mechanical Engineering

This report documents all the research, ideation, and mockups used to determine right ventricle motion and develop a system capable of reproducing that motion on a tissue sample. The model is intended for evaluating anchoring systems being developed by Edwards Lifesciences for use with tricuspid valve therapies. Several design solutions were considered for the primary functions of recreating motion of the right ventricle and attaching tissue to the device. From these ideas a primary means of producing motion and attaching tissue was selected. These ideas were then developed over the course of a school year to become the final system …

Autonomous Navigation And Mapping Using Lidar, Steven E. Alsalamy, Ben C. Foo, Garrett C. Frels

Computer Engineering

The goal of this project was to make a fully autonomous robot, capable of mapping its surroundings and navigating through obstacles. This was done through the use of a chassis fitted with tracks and two motors, a lidar, a compass, and a Raspberry Pi. The robot also contains two batteries and is self powered. Encoders are used on the motors in order to track the distance traveled for more precise mapping and movements.

Darling, Robot For Roborodentia 2018, Michael Le, Steven Liu

Computer Engineering

For our senior project, our group decided to build a robot to participate in Roborodentia 2018, an annual robotics competition overlooked by Professor Seng that takes place during open house. When taking into consideration the classes that Computer Engineering students had to have taken and the skills that we have developed throughout our time here on campus, a robotics project seemed to be an appropriate culmination of both the technical and non-technical skills that we have acquired.

Roborodentia, Bryan D. Hendricks

Computer Engineering

This project is an autonomous robot, designed to perform a series of basic tasks without any human input. It’s based on the 2018 Roborodentia competition, in which teams of students design and build a small (roughly 1 square foot) robot that collects small foam spheres from vertical tubes on the edges of a table-sized arena, and shoot them into goals across the field. The more foam spheres the robot makes into the goals after a 3 minute time period, the more points they get. The challenge is doing so autonomously, without any human input after the initial timer for the …

Roborodentia Final Report, Trevor James Gesell, Zeph Colby Nord, Mitchell Tyler Myjak

Computer Engineering

The Senior Project consisted of competing in Roborodentia, a competition in which groups build robots to complete a particular task. This event took place at the Cal Poly Open House on Saturday, April 12th, 2018. For the competition, the robot was to collect Nerf balls from supply tubes raised approximately 7” from the board and shoot them into nets placed along the opposite side of the course. The design, manufacture, and testing of the robot began the first week of Cal Poly winter quarter and lasted until the day of the competition.

A Comparative Study Of Feature Detection Methods For Auv Localization, Andrew Y. Kim

Master's Theses

Underwater localization is a difficult task when it comes to making the system autonomous due to the unpredictable environment. The fact that radio signals such as GPS cannot be transmitted through water makes autonomous movement and localization underwater even more challenging. One specific method that is widely used for autonomous underwater navigation applications is Simultaneous Localization and Mapping (SLAM), a technique in which a map is created and updated while localizing the vehicle within the map. In SLAM, feature detection is used in landmark extraction and data association by examining each pixel and differentiating landmarks pixels from those of the …

Artificial Neural Network-Based Robotic Control, Justin Ng

Master's Theses

Artificial neural networks (ANNs) are highly-capable alternatives to traditional problem solving schemes due to their ability to solve non-linear systems with a nonalgorithmic approach. The applications of ANNs range from process control to pattern recognition and, with increasing importance, robotics. This paper demonstrates continuous control of a robot using the deep deterministic policy gradients (DDPG) algorithm, an actor-critic reinforcement learning strategy, originally conceived by Google DeepMind. After training, the robot performs controlled locomotion within an enclosed area. The paper also details the robot design process and explores the challenges of implementation in a real-time system.

Corridor Navigation For Monocular Vision Mobile Robots, Matthew James Ng

Master's Theses

Monocular vision robots use a single camera to process information about its environment. By analyzing this scene, the robot can determine the best navigation direction. Many modern approaches to robot hallway navigation involve using a plethora of sensors to detect certain features in the environment. This can be laser range finders, inertial measurement units, motor encoders, and cameras.

By combining all these sensors, there is unused data which could be useful for navigation. To draw back and develop a baseline approach, this thesis explores the reliability and capability of solely using a camera for navigation. The basic navigation structure begins …

Automated Pruning Of Greenhouse Indeterminate Tomato Plants, Joey M. Angeja

Master's Theses

Pruning of indeterminate tomato plants is vital for a profitable yield and it still remains a manual process. There has been research in automated pruning of grapevines, trees, and other plants, but tomato plants have yet to be explored. Wage increases are contributing to the depleting profits of greenhouse tomato farmers. Rises in population are the driving force behind the need for efficient growing techniques. The major contribution of this thesis is a computer vision algorithm for detecting greenhouse tomato pruning points without the use of depth sensors. Given an up-close 2-D image of a tomato stem with the background …

Real Time And High Fidelity Quadcopter Tracking System, Tyler Mckay Hall

Computer Engineering

This project was conceived as a desired to have an affordable, flexible and physically compact tracking system for high accuracy spatial and orientation tracking. Specifically, this implementation is focused on providing a low cost motion capture system for future research. It is a tool to enable the further creation of systems that would require the use of accurate placement of landing pads, payload acquires and delivery. This system will provide the quadcopter platform a coordinate system that can be used in addition to GPS.

Field research with quadcopter manufacturers, photographers, agriculture and research organizations were contact and interviewed for information …

Remote Cable Gantry, Allen L. Bailey

Electrical Engineering

The Remote Cable Gantry is a robotic system that was initially intended to aid in the art of aerial videography. It was designed to enable novice and expert users alike to capture both video footage and audio from perspectives unachievable by current methods. This system uses a series of cables to control the position of a camera gimbal in a defined 3D space and, as a self-contained unit, is portable and easy to use. The Remote Cable Gantry offers a quiet, intuitive, and safe alternative to existing technology, which has been limiting the market and potential of aerial photography and …

Light Sensing Automated Blinds, Andrew B. Hodges, Ryan C. Flick

Electrical Engineering

This project is solving the problem of wasted energy within buildings and homes, because currently the lights turned on inside building do not utilize natural, ambient light from the sun. Rather than having unnecessary light from a light source, the automated light sensing smart blinds can sense the amount of light outside the window and in the room, and then adjust the angle of the blinds to save energy by utilizing the available outdoor light. This way, the light source will not be running at maximum power output while there is excess light coming through the window. This project aims …

The Following Robot, Juan D. Cerda, Matthew S. Kwan, Vi M. Le

Computer Engineering

The objective of this project is to design, build, and test an autonomous robot with an associated Android application. The robot uses on board inertial measurement sensors (magnetometer, accelerometer, gyroscope) and coordinates itself through Bluetooth communication with the similar built­in measurement sensors on the Android phone to mimic and follow movement. The Following Robot incorporates the same basic movement functionality as a typical RC car. The robot follows the user’s phone through an application on one’s phone. This application accesses the phone’s accelerometer and gyroscope data and translates into appropriate conversions. Methods of tracking and calculating distance or angular displacement …

Modeling Autonomous Vehicles Through Radio Controlled Cars, Eva S. Chen

Computer Engineering

Autonomous vehicles have a lot of potential in improving people’s everyday lives. They could reduce congestion, reduce collisions, enhance mobility, and more. But with these benefits come security and privacy risks. In order to research and test some of these risks, we are building a set of scale autonomous cars that can model autonomous and collaborative behaviors. One such behaviour would be platooning, where a group of vehicles can travel closely together at high speeds by following a lead car. We are doing this with various sensors and control algorithms to allow for future modularity.

Farmbot Rfid Integration, Laura R. Swart

Computer Engineering

The purpose of this project is to assist the company FarmBot improve their product by adding RFID tracking to the FarmBot robot. RFID tracking will allow the robot to select and pick up tool heads without any user interference.

Blend It Wine Blending Distribution System, Connor Clarry, Russell Temple, Matt Moren

Mechanical Engineering

No abstract provided.

Wheelieking Trainer Project Report, Harold Hall Iii, Nathaniel Fox, Thomas Niemisto

Biomedical Engineering

In this report we will detail the design and implementation process of the WheelieKing Trainer project, a device that helps people learn how to do wheelies on a bicycle by preventing backward falls. Formal project requirements are specified, followed by the ideation and iteration process to meet those requirements. The components and methods used to create the device are described in detail. The results of the development process and usage test results of the device are included. Appendices at the end of this report include references, supporting analyses, and project management and timeline details.

Models For Pedestrian Trajectory Prediction And Navigation In Dynamic Environments, Jeremy N. Kerfs

Master's Theses

Robots are no longer constrained to cages in factories and are increasingly taking on roles alongside humans. Before robots can accomplish their tasks in these dynamic environments, they must be able to navigate while avoiding collisions with pedestrians or other robots. Humans are able to move through crowds by anticipating the movements of other pedestrians and how their actions will influence others; developing a method for predicting pedestrian trajectories is a critical component of a robust robot navigation system. A current state-of-the-art approach for predicting pedestrian trajectories is Social-LSTM, which is a recurrent neural network that incorporates information about neighboring …

Software Updates To A Multiple Autonomous Quadcopter Search System (Maqss), Jared Speck, Toby Chan

Computer Engineering

A series of performance-based and feature implementation software updates to an existing multiple vehicle autonomous target search system is outlined in this paper. The search system, MAQSS, is designed to address a computational power constraint found on modern autonomous aerial platforms by separating real-time and computationally expensive tasks through delegation to multiple multirotor vehicles. A Ground Control Station (GCS) is also described as part of the MAQSS system to perform the delegation and provide a low workload user interface. Ultimately, the changes to MAQSS noted in this paper helped to improve the performance of the autonomous search mission, the accuracy …

Animatronic Kinect Bear, Christopher J. Barth, Luis Manjarrez, Emily Lopez

Computer Engineering

The objective of the project was to build an interactive robotic bear through the application of Computer Engineering related software and hardware concepts. Utilizing the body recognition features of XBos One Kinect, the bera will be able to mimic a user's motions in real time.

Robostock: Autonomous Inventory Tracking, Drew Christian Balthazor

Computer Engineering

No abstract provided.

Android Drone: Remote Quadcopter Control With A Phone, Aubrey John Russell

Computer Engineering

The purpose of the “Android Drone” project was to create a quadcopter that can be controlled by user input sent over the phone’s Wi-Fi connection or 4G internet connection. Furthermore, the purpose was also to be able to receive live video feedback over the internet connection, thus making the drone an inexpensive option compared to other, equivalent drones that might cost thousands of dollars. Not only that, but the Android phone also has a host of other useful features that could be utilized by the drone: this includes GPS, pathing, picture taking, data storage, networking and TCP/IP, a Java software …

Autonomous Golf Cart Firmware, Gerik Kubiak

Computer Engineering

The Autonomous Golf Cart Project is a project sponsored by the Cal Poly Robotics Club. The multidisciplinary team is adding sensors and electronics to a regular golf cart with the goal to drive the golf cart around campus without and human input. This task requires a plethora of hardware and firmware to control that hardware. The firmware provides an interface for higher level software to then control the hardware and therefore drive the golf cart. This report is focused on the hardware modifications and the firmware used in order to drive the golf cart from a computer.