Search papers, labs, and topics across Lattice.
This paper compares Reinforcement Learning (RL) and Nonlinear Model Predictive Control (NMPC) for energy-aware path following in electric vehicles, incorporating explicit regenerative braking within a Frenet frame kinematic model. The study evaluates NMPC (with JIT-compiled CasADi), Proximal Policy Optimization (PPO), a PID baseline, and a Stanley geometric baseline. The key result shows that the trained PPO policy demonstrates significant transferability to diverse and unseen tracks, even to a more complex dynamic single-track model with minimal fine-tuning, outperforming other controllers in energy efficiency.
A single RL policy masters energy-efficient path following across diverse tracks and even complex vehicle dynamics, outperforming NMPC and baselines without explicit track-specific training.
Path-following control strategies typically follow the bi-objective optimization dilemma: minimizing deviations from a reference path while maintaining smooth speed profiles. The latter objective is especially relevant for Electric Vehicles (EVs), since their limited driving range can be extended by recovering energy through regenerative braking, a feature that has not yet been sufficiently studied in the literature. In this work, we perform a comparative analysis of four controllers under one common Frenet frame-based kinematic vehicle model, utilizing a validated energy model (VT-CPEM) with explicit regenerative braking. Herein, we implement the following controllers: Nonlinear Model Predictive Control (NMPC), Proximal Policy Optimization (PPO), gain-scheduled Ackermann state-feedback baseline (PID-SF), and a Stanley geometric baseline. To satisfy real-time requirements, we implement the NMPC using JIT-compiled CasADi. Moreover, we train the PPO using traditional straight and S-curve tracks, after which we successfully transfer the unmodified policy to unseen tracks, including: an ISO 3888-1 lane-change, a chicane, randomly-generated parameterized-splines, and a $\pm3^\circ$ graded road. In addition, the policy transfers to a dynamic single-track vehicle model with linear tires, zero-shot with an acceptable initial performance, which was optimized after brief fine-tuning. Thereby, we demonstrate that our PPO is readily transferable to more comprehensive vehicle models. We conclude with a performance analysis of developed controllers and discuss ideas for future work.
