Dipam Patel
Graduate Student
Graduate Research Assistant
Joined department: Spring 2023
Education
Dipam is a Ph.D. student in the Department of Computer Science at Purdue University. He is mainly interested in Deep Learning & Computer Vision and focuses on developing optimized algorithms to run on embedded devices. His work involves giving intelligent vision to aerial & ground robots for real-time decision-making and behavioral understanding.
Before joining Purdue, he completed his Masters’s in Robotics from the University of Maryland, College Park, and his Bachelor’s in Mechanical Engineering from Gujarat Technological University, India. After his Master, he worked as a Robotics Software Engineer for three years at Airgility - An Unmanned Autonomous Vehicles Startup, where he was one of the earliest employees. At Airgility, he also filed his first patent titled ‘System and Method for Enabling Efficient Hardware-to-Software Performance’.
Dipam is passionate to dive deeper into the fields of Computer Vision, Deep Learning, and Augmented Reality. He is interested in developing and implementing sophisticated algorithms to create an impactful outcome for the community through the use of Next Generation Robotics.
Selected Publications
DroNeRF: Real-time Multi-agent Drone Pose Optimization for Computing Neural Radiance Fields [Paper]
Dipam Patel, Phu Pham, Aniket Bera | 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
We present a novel optimization algorithm called DroNeRF for the autonomous positioning of monocular camera drones around an object for real-time 3D reconstruction using only a few images. Neural Radiance Fields, or NeRF, is a novel view synthesis technique used to generate new views of an object or scene from a set of input images. Using drones in conjunction with NeRF provides a unique and dynamic way to generate novel views of a scene, especially with limited scene capabilities of restricted movements. Our approach focuses on calculating optimized pose for individual drones while solely depending on the object geometry without using any external localization system. The unique camera positioning during the data capturing phase significantly impacts the quality of the 3D model. To evaluate the quality of our generated novel views, we compute different perceptual metrics like the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). Our work demonstrates the benefit of using an optimal placement of various drones with limited mobility to generate perceptually better results.
Scaling Safe Multi-Agent Control for Signal Temporal Logic Specifications
Joe Eappen, Zikang Xiong, Dipam Patel, Aniket Bera, Suresh Jagannathan | 2024 Conference on Robot Learning (CoRL)
Existing methods for safe multi-agent control using logic specifications like Signal Temporal Logic (STL) often face scalability issues. This is because they rely either on single-agent perspectives or on Mixed Integer Linear Programming (MILP)-based planners, which are complex to optimize. These methods have proven to be computationally expensive and inefficient when dealing with a large number of agents. To address these limitations, we present a new scalable approach to multi-agent control in this setting. Our method treats the relationships between agents using a graph structure rather than in terms of a single-agent perspective. Moreover, it combines a multi-agent collision avoidance controller with a Graph Neural Network (GNN) based planner, models the system in a decentralized fashion, and trains on STL-based objectives to generate safe and efficient plans for multiple agents, thereby optimizing the satisfaction of complex temporal specifications while also facilitating multi-agent collision avoidance. Our experiments show that our approach significantly outperforms existing methods that use a state-of-the-art MILP-based planner in terms of scalability and performance.