Continuous measurement improves prediction of vehicle server delays
Closing the Loop: Continuous Measurement-Driven Refinement of Offloading Predictions
Software Engineering
Summary
Vehicles often send heavy computing tasks to powerful backend servers, but predicting how long this takes can be tricky because of changing network and server conditions. The authors show that by constantly measuring real delays and updating their prediction model during operation, they can better keep track of actual performance. This approach helps reduce errors in estimating delay times and server loading, although some unpredictability remains due to complex and variable conditions. Overall, this ongoing adjustment is needed to make vehicle offloading more reliable in changing environments.
What this means in practice
- •For vehicle software teams: Maintain accurate latency predictions for cloud-based vehicle perception and control by continuously updating prediction models with live performance data.
- •For cloud infrastructure operators: Use ongoing feedback from running applications to refine performance predictions across heterogeneous hardware and dynamic load conditions for better resource management.
Authors
Falk Dettinger, Matthias Weiß, Michael Weyrich
Abstract
Modern vehicles increasingly offload computation- ally intensive perception and decision functions to backend servers, requiring accurate predictions of absolute performance metrics such as Round-Trip Time (RTT), processing time, and utilization. In practice, strong temporal variability, heterogeneous backend hardware, and multimodal latency regimes cause offline- trained predictors to drift, creating a reliability gap for latency- sensitive functions. We address this gap with an operational, measurement-driven closed loop that continuously recalibrates absolute-value predictors during runtime. The system aligns real execution measurements with predicted values and performs incremental online updates of a lightweight multi-head neural network while preserving model stability. The model implicitly learns the broad, non-Gaussian spread of input metrics, and a sigma-based error analysis in our evaluation characterizes resid- ual variability under dynamic conditions. Experiments across two Kubernetes clusters show that continuous measurement- driven refinement reduces prediction drift, improves accuracy for RTT, processing time, and utilization, and stabilizes prediction behavior across heterogeneous latency regimes. However, the broad and multimodal distribution of input metrics imposes fundamental limits on absolute-value prediction, with residual errors frequently exceeding configured thresholds. Overall, online calibration proves feasible and necessary for robust computation offloading in dynamic vehicular edge environments, while high- lighting the need for future mechanisms that address extreme latency regimes and high-variance operating conditions.