Visual Navigation for Autonomous Material Deposition Systems Using Remote Sensing

• Author / Creator

  Maleki, Soroush

• In this work, a novel visual navigation method is proposed to estimate the state of mobile and fixed cold-spray material deposition systems using a stereocamera sensor installed in the workspace. Unlike other visual localization algorithms that exploit costly onboard sensors such as LiDARs or fully rely on distinct visual cues on the robot and grid markers in the environment, our method significantly reduces the cost and complexity of the sensory setup by utilizing a cost-effective remote stereo vision system. This allows for the localization of the target system regardless of its appearance or the environment and enables scalability for tracking and operation of multiple mobile material deposition systems at the same time. To achieve this aim, deep neural networks, kinematic constraints, and learning-aided state observers are employed to detect and estimate the location and orientation of the deposition system. A physical model of the system is fused with a remote visual sensing module and is proposed. This accounts for frames in which depth estimation accuracy is reduced due to perceptually degraded conditions in the cold spraying context. The visual state estimation algorithm is evaluated on a fixed and mobile setup that demonstrates the accuracy and reliability of the proposed method. Moreover, a model predictive controller is formulated, designed, and implemented to enable the task of autonomous mobile robot trajectory tracking. The architecture of the model predictive controller incorporates essential kinematic constraints, input bounds, and control input smoothness, thereby ensuring the generation of a feasible input for the autonomous mobile material deposition system to seamlessly trace a predefined trajectory. A comprehensive comparative analysis is conducted between this model predictive controller and its PID counterpart, encompassing a rigorous evaluation through a series of simulated and real-world tests, aiming to elucidate their respective performances and characteristics.

• Subjects / Keywords

  • Robotics

• Graduation date

  Fall 2023

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-ft2v-bd82

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.