LLM Adaptive PID Control for B5G Truck Platooning Systems

Abstract

This paper presents an exploration into the capabilities of an adaptive PID controller within the realm of truck platooning operations, situating the inquiry within the context of Cognitive Radio and AI-enhanced 5G and Beyond 5G (B5G) networks. We developed a Deep Learning (DL) model that emulates an adaptive PID controller, taking into account the implications of factors such as communication latency, packet loss, and communication range, alongside considerations of reliability, robustness, and security. Furthermore, we harnessed a Large Language Model (LLM), GPT-3.5-turbo, to deliver instantaneous performance updates to the PID system, thereby elucidating its potential for incorporation into AI-enabled radio and networks. This research unveils crucial insights for augmenting the performance and safety parameters of vehicle platooning systems within B5G networks, concurrently underlining the prospective applications of LLMs within such technologically advanced communication environments.

Keywords: 5G and B5G systems; V2V communication; adaptive PID control; coordination of vehicles; large language models; platooning.

Figure 1

Flow diagram of the adaptive PID controller with LLM performance updates.

Figure 2

Training and validation loss over time for the proposed architecture.

Figure 3

Detailed network architecture of the deep neural network (DNN) used for the adaptive PID tuning. The model consists of two fully connected layers with 64 and 32 neurons, followed by dropout layers with a rate of 0.2. Rectified Linear Unit (ReLU) activation functions, depicted as $\sigma$, are applied after each fully connected layer. The input and output layers are also depicted.

Figure 4

Desired vs. actual inter-vehicle distance with latency.

Figure 5

Temporal evolution of the control signal amid the latency. The control signal encapsulates the system modifications applied to sustain the requisite distance between the vehicles within a truck platoon, acting as a responsive adjustment to the latency-induced variations.

Figure 6

Desired vs. actual inter-vehicle distance with packet loss.

Figure 7

Control signal trajectory amid the packet loss. This illustrates the control signal’s role as a corrective mechanism that dynamically adjusts to maintain the intended inter-vehicle distance within a truck platoon, demonstrating its resilience despite the packet loss events.

Figure 8

Sigmoid function for gradual packet loss rate. This plot illustrates the relationship between the distance ratio (predicted distance divided by communication range) and the packet loss rate. The sigmoid function demonstrates a gradual increase in packet loss rate as the distance ratio increases, simulating a more realistic communication scenario in the control loop.

Figure 9

Desired vs. actual inter-vehicle distance with a gradual packet loss.

Figure 10

Temporal progression of the control signal amid gradual packet loss. The control signal functions as an adaptive mechanism that continually adjusts to preserve the targeted distance between vehicles within a truck platoon, even when confronting the challenges of a gradual packet loss.

Figure 11

Desired vs. actual inter-vehicle distance with the communication range.

Figure 12

Control signal trajectory in varying communication ranges. The control signal, depicted here, acts as a real-time corrective measure that effectively regulates inter-vehicle distance within a truck platoon, demonstrating its adaptability across different communication range scenarios.

Figure 13

Desired vs. actual inter-vehicle distance with noisy communication.

Figure 14

Control signal behavior amid noisy communication. This depiction of the control signal underscores its role as a dynamic corrective measure, adjusting in real time to manage inter-vehicle distances within a truck platoon, even under the challenging conditions of communication noise.

Figure 15

Desired vs. actual inter-vehicle distance with encrypted communication.

Figure 16

Control signal over time with encrypted communication. The control signal represents the adjustment applied to the system to maintain the desired distance between the vehicles in a truck platoon.