面向高能效和低时延的智能驾驶任务调度方法及装置
By constructing a pre-defined scheduling model using deep reinforcement learning and Markov decision process theory, the scheduling of autonomous driving tasks in local or roadside units is optimized, solving the problems of vehicle power consumption and poor wireless channel conditions, and achieving high energy efficiency and low latency task processing.
CN121900842BActive Publication Date: 2026-07-17北京智慧城市网络有限公司
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 北京智慧城市网络有限公司
- Filing Date
- 2025-12-29
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing autonomous driving task scheduling strategies fail to transmit data during peak hours when vehicle battery is consumed too quickly or when wireless channel conditions are poor, affecting driving range and task real-time performance.
Method used
A pre-defined scheduling model, constructed using deep reinforcement learning and constrained Markov decision process theory framework, monitors vehicle status and communication environment in real time, optimizing task scheduling decisions at the local or roadside unit level.
Benefits of technology
It extends the vehicle's driving range, improves computing and power resource utilization efficiency, and ensures the reliability and real-time performance of task transmission.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN121900842B_ABST
Abstract
本申请公开了一种面向高能效和低时延的智能驾驶任务调度方法及装置，方法包括：监测待处理驾驶任务，确定待处理驾驶任务的任务属性；实时监测车辆的电量状态和车载计算资源状态作为车辆状态信息；通过无线通信模块监测无线信道状态，以获取当前的通信环境信息；将任务属性、车辆状态信息、通信环境信息输入预设调度模型中进行处理，输出待处理驾驶任务对应的调度指令；当调度指令指示本地计算时，将待处理驾驶任务输入车载计算单元；或者在调度指令指示车路协同时，将待处理驾驶任务卸载至路侧单元。因此，采用本申请实施例，可避免在无线信道状态不佳时的高功耗传输，有效提高计算和电力资源的利用效率，并同时确保任务传输的可靠性和实时性。
Need to check novelty before this filing date? Find Prior Art