SARS: A Resource Selection Algorithm for Autonomous Driving Tasks in Heterogeneous Mobile Edge Computing

Reza Zakerian,Hadi Gholami

2024-11-25

Abstract:With the rapid advancement of devices requiring intensive computation, such as Internet of Things (IoT) devices, smart sensors, and wearable technology, the computational demands on individual platforms with limited resources have escalated, necessitating the offloading of the generated tasks by the devices to edge. These tasks are often real-time with strict response time requirements. Among these devices, autonomous vehicles present unique challenges due to their critical need for timely and accurate processing to ensure passenger safety. Selecting suitable servers in a heterogeneous mobile edge computing (MEC) architecture is vital to optimizing real-time task processing rates for such applications. To address this, we present an algorithmic solution to improve the allocation of heterogeneous servers to real-time tasks, aiming to maximize the number of processed tasks. By analyzing task and server characteristics in the MEC architecture, we develop the suitability-based adaptive resource selection (SARS) algorithm, which evaluates server suitability based on factors like time constraints and server capabilities. Additionally, we introduce the proactive on-demand resource allocation (PORA) algorithm, which strategically reserves computational resources to ensure availability for critical real-time tasks. We compare the proposed algorithms with several classical and state-of-the-art algorithms. Computational results demonstrate that our approach outperforms existing algorithms, processes more tasks, and effectively prioritizes urgent tasks, particularly in autonomous driving applications.

Distributed, Parallel, and Cluster Computing,Networking and Internet Architecture

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to select an appropriate server for autonomous driving tasks in a heterogeneous Mobile Edge Computing (MEC) architecture in order to optimize the real - time task processing rate. Specifically, the paper focuses on how to maximize the number of processed tasks under limited resources and ensure that tasks are completed within the specified time, especially in time - sensitive autonomous driving applications.

### Problem Background

With the rapid development of devices that require a large amount of computing resources, such as Internet of Things (IoT) devices, smart sensors, and wearable technologies, the computing resources of a single platform are becoming increasingly strained, causing the tasks generated by these devices to need to be offloaded to Edge Computing Servers (ECS). Especially for autonomous vehicles, due to their strict requirements for timeliness and accuracy, it is crucial to select an appropriate server to process these tasks. Autonomous vehicles need to process a large amount of data from sensors such as radars, LiDARs, and cameras, and perform computationally intensive tasks such as object detection, path planning, and sensor fusion. Any delay may lead to catastrophic consequences.

### Specific Problems

1. **Timeliness of Task Offloading**: Autonomous driving tasks are time - critical and must be completed within the specified time. If these time constraints are not met, the tasks will become invalid, affecting the safety and reliability of the vehicle.

2. **Efficiency of Resource Allocation**: In a heterogeneous MEC architecture, different servers have different computing capabilities. How to efficiently allocate these resources to ensure that tasks can be completed in the shortest time while avoiding resource waste and overload is an important challenge.

3. **Limitations of Existing Algorithms**: Most existing Resource Selection Algorithms (RSP) rely on simple rules, such as selecting the server with the shortest execution time and waiting time, which may lead to sub - optimal performance and low resource utilization in a heterogeneous environment.

### Solutions

To address the above challenges, the paper proposes two new algorithms:

1. **Suitability - based Resource Selection Algorithm (SARS)**:
- SARS evaluates the suitability of each server by comprehensively considering factors such as task characteristics, time constraints, and server capabilities, thereby optimizing resource utilization and ensuring a high task completion rate.
- The SARS algorithm traverses all available Processing Units (PU), calculates the suitability score of each PU, and selects the PU that is most suitable for processing the current task.

2. **Proactive On - Demand Resource Allocation Algorithm (PORA)**:
- PORA ensures that resources can be immediately allocated for processing when an urgent task arrives by reserving some Processing Units (PU) with strong computing capabilities in advance, thereby improving the system's response speed and efficiency.
- When SARS cannot find a suitable PU, PORA will activate the reserved PU to process the emergency task and then restore it to the standby state.

### Experimental Results

The experimental results show that the SARS and PORA algorithms are significantly superior to existing algorithms in terms of task completion rate, especially when processing emergency tasks. By adjusting the parameters α and β, the SARS algorithm can more effectively balance the time margin and load factors, further improving the efficiency of task processing.

In conclusion, this paper aims to improve the processing efficiency and reliability of autonomous driving tasks in a heterogeneous MEC architecture by improving resource selection and allocation strategies, ensuring that more tasks can be completed within the specified time.