CRII: CSR: Energy-Aware Coordination and Management of Multi-Purpose Autonomous Robots for Maintained Continuity of Operations and Long Battery Lifetime
NSF Award Page: Link
PROJECT ABSTRACT — Autonomous Robot (AR) fleets will soon assist humans for the execution of several tasks. One of the most important factors that will likely affect the widespread adoption of AR fleets in our society is their ease of management for the end users. The goal of this project is to achieve the easy management of the AR fleet by allowing end users to indicate the set of tasks to execute during a working period and by implementing algorithms that automatically (i) allocate tasks to ARs, (ii) coordinate recharge schedules, (iii) ensure continuity of operations, and (iv) minimize the battery degradation.
The intellectual merits of this project are focused on designing an AR fleet manager system that can be used by end users to interact easily with an AR fleet. Three main research thrusts are: (i) joint optimization of concurrent tasks allocation and recharge scheduling for ARs to minimize the task downtime and battery degradation during the working period; (ii) coordination of the ARs’ computing and sensing resources to balance tasks performance and energy consumption; (iii) full system implementation and experimentation on a small-scale physical testbed and a large-scale simulated testbed.
The proposed work will introduce new technologies available for the management of AR fleets and boost the widespread use of autonomous robots in our society, e.g., homes, smart farms, industry, and smart cities. The project will provide a new hardware and software testbed to help train and inspire undergraduate and graduate students. New industry collaborations will be established to enhance AR adoption. New academic collaborations will also be established with researchers in various areas including robotics, artificial intelligence, control systems, sensing, and approximation algorithms.
Publications and Artifacts Related to This Award
- Syeda Tanjila Atik, Akshar Shravan Chavan, Daniel Grosu, and Marco Brocanelli, “A Maintenance-Aware Approach for Sustainable Autonomous Mobile Robot Fleet Management“, IEEE Transactions on Mobile Computing, 2023 (Accepted To Appear)
MTC Artifact: Github Repository
- Akshar Chavan, and Marco Brocanelli, “Towards High-Quality Battery Life for Autonomous Mobile Robot Fleets“, IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS), 2022
TCM Artifact: Github Repository
- Tayebeh Bahreini, Marco Brocanelli, and Daniel Grosu, “VECMAN: A Framework for Energy-Aware Resource Management in Vehicular Edge Computing Systems“, IEEE Transactions on Mobile Computing, 2021 (Accepted To Appear)
PAPER ABSTRACT — In Vehicular Edge Computing (VEC) systems, the computing resources of connected Electric Vehicles~(EV) are used to fulfill the low-latency computation requirements of vehicles. However, local execution of heavy workloads may drain a considerable amount of energy in EVs. One promising way to improve the energy efficiency is to share and coordinate computing resources among connected EVs. However, the uncertainties in the future location of vehicles make it hard to decide which vehicles participate in resource sharing and how long they share their resources so that all participants benefit from resource sharing. In this paper, we propose VECMAN, a framework for energy-aware resource management in VEC systems composed of two algorithms: (i) a resource selector algorithm that determines the participating vehicles and the duration of resource sharing period; and (ii) an energy manager algorithm that manages computing resources of the participating vehicles with the aim of minimizing the computational energy consumption. We evaluate the proposed algorithms and show that they considerably reduce the vehicles computational energy consumption compared to the state-of-the-art baselines. Specifically, our algorithms achieve between 7% and 18% energy savings compared to a baseline that executes workload locally and an average of 13% energy savings compared to a baseline that offloads vehicles workloads to RSUs.
Graduate Research Training Funded by this Award
- Akshar Chavan – Graduate Research Assistant, Fall 2020 – Winter 2022
Undergraduate Research Training in Autonomous Mobile Robots
- Anmol Multani
Honors Thesis Title: “Simulating the Energy Consumption of Autonomous Mobile Robots (AMRs)” - Nitisha Omkar
Honors Thesis Title: “Simulating the State of Charge of Autonomous Mobile Robots (AMRs) Using Optimization Technologies” - Sam Walsh
Research Experience for Undergraduates Site (Summer Academy in Sustainable Manufacturing): “Optimization of battery life and energy usage for autonomous robots”
Curriculum Development
- CSC 4996 (2021-2023) – Special Capstone in Robotics: In collaboration with Dr. Abhilash Pandya, we have created a section of robotics branching from the original capstone project class. This program allows students participating in the existing robotics club of WSU to get undergraduate credits. The program is designed to span over two semesters and allows undergraduate students to get familiar with the Robotic Operating System (ROS), participate in the software/hardware implementation of a real autonomous mobile robot, and participate in the annual Intelligent Ground Vehicle Competition (IGVC). Robotics Club Website: Link, IGVC Competition Website: Link
- CSC 8260 (Winter 2021) – Ph.D.-oriented seminar to focus on energy efficiency in edge computing. We studied recently published papers in hot-topics including IoT devices, Machine Learning at the Edge, autonomous vehicles, and autonomous mobile robots. Each student conducts a research project, which will be submitted to a conference. In addition, each student had to conduct several class presentations to help them improve their presentation skills.
- CSC 8260 (Winter 2022) – Ph.D.-oriented seminar to focus on managing tasks and resources in modern mobile/edge/cloud systems. Specifically, we study papers published in 2021 MobiSys, MobiCom, ISCA, ATC, EuroSys, and ASPLOS focusing on hot-topics such as: parallelization of model inference on mobile GPU and allocation to edge/cloud, inference time prediction on heterogeneous edge systems, memory usage optimization for CPU inference, concurrent training on GPU, decision trees, augmented and virtual reality optimization, caching, prefetching, energy-aware adaptation of mobile apps, multi-device app execution, task allocation in shared vehicles (e.g., drones, delivery robots), CPU usage optimization through learning-based DVFS and idle CPU harvesting, multi-task and multi-processor resource partitioning, datacenter overclocking and oversubscription. Each student must conduct a novel research project, which will be submitted to a conference after the end of the semester. In addition, each student must conduct class presentations and learn how to properly critique research papers.