A-Priori Computation Estimation in Fog Computing Robotics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile robotic and drone nodes are restricted by limited computation and battery power, making it challenging to perform on-board runtime computations effectively, and existing offloading methods to cloud or robotic peers do not adequately address latency and energy efficiency in dynamic environments.
Innovation Solution
A method and system for dynamically profiling computational tasks on a development testbed to estimate a-priori computational time and energy requirements, extrapolating these on heterogeneous deployment hardware using benchmarks and parallel processing models, and intelligently allocating tasks based on pre-defined rules considering computational complexity, latency, and processing power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If computational tasks are offloaded to cloud or robotic peers, then computation capacity is improved, but latency and energy efficiency are not adequately addressed
Solution Approach 1:
The patent introduces fog nodes as intermediary computing resources between mobile robotic nodes and cloud infrastructure. These fog nodes provide localized computational power that reduces communication latency compared to cloud offloading, while maintaining the ability to handle complex tasks that exceed onboard robot capabilities. The system dynamically selects fog nodes as intermediaries to balance computation capacity and latency requirements.
Solution Approach 2:
The system performs a-priori estimation of execution times and energy depletion rates for offloaded computational programs before actual execution. This preliminary analysis enables the robotic nodes to predict performance outcomes and make informed decisions about task offloading, selecting optimal fog nodes in advance to minimize latency and energy consumption.
2Power
If computational tasks are offloaded to cloud or robotic peers, then computation capacity is improved, but energy efficiency deteriorates
Solution Approach 1:
The patent implements a hierarchical offloading strategy where computational tasks are routed to locally available fog nodes rather than distant cloud data centers. This localizes computation resources closer to the robotic nodes, reducing communication energy costs and improving overall energy efficiency. The system evaluates energy depletion rates and selects fog nodes that minimize energy consumption while providing sufficient computational capacity.
Solution Approach 2:
The system performs a-priori estimation of energy depletion rates for offloaded computational programs before actual execution. This preliminary energy analysis enables the robotic nodes to predict energy consumption outcomes and make informed decisions about task offloading, selecting optimal fog nodes in advance to minimize energy usage while maintaining computation capacity.
3Productivity
If a-priori estimation of execution times is implemented, then task allocation efficiency is improved, but system complexity increases
Solution Approach 1:
The patent employs profiling techniques that create simplified models or copies of computational task characteristics and fog node capabilities. Instead of implementing complex real-time analysis, the system uses pre-collected profiling data to estimate execution times and energy requirements. This copying approach maintains task allocation efficiency while reducing the computational complexity required during actual runtime decision-making.
Data Source
AI summary
In order to make use of computational resources available at runtime through fog networked robotics paradigm, it is critical to estimate average performance capacities of deployment hardware that is generally heterogeneous. It is also not feasible to replicate runtime deployment framework, collected sensor data and realistic offloading conditions for robotic environments. In accordance with an embodiment of the present disclosure, computational algorithms are dynamically profiled on a development testbed, combined with benchmarking techniques to estimate compute times over the deployment hardware. Estimation in accordance with the present disclosure is based both on Gustafson's law as well as embedded processor benchmarks. Systems and methods of the present disclosure realistically capture parallel processing, cache capacities and differing processing times across hardware.