Organizations: Department of Electronics, Information and Bioengineering (DEIB), Politecnico di Milano, Milan, Italy · Department of Applied Mathematics, Ferdowsi University of Mashhad, Mashhad, Iran · Department of Applied Mathematics, Faculty of Mathematical Sciences, University of Guilan, Rasht, Iran · Department of Mechanical Engineering, Politecnico di Milano, Milan, Italy · Department of Robotics and Control Engineering, Shahrood University of Technology, Shahrood, Iran
Abstract
The integration of Model Predictive Control (MPC) and Reinforcement Learning (RL) has emerged as a promising paradigm for constrained decision-making and adaptive control. MPC offers structured optimization, explicit constraint handling, and established stability tools, whereas RL provides data-driven adaptation and performance improvement in the presence of uncertainty and model mismatch. Despite the rapid growth of research on RL--MPC integration, the literature remains fragmented, particularly for control architectures built on linear or linearized predictive models. This paper presents a comprehensive Systematic Literature Review (SLR) of RL--MPC integrations for linear and linearized systems, covering peer-reviewed and formally indexed studies published until 2025. The reviewed studies are organized through a multi-dimensional taxonomy covering RL functional roles, RL algorithm classes, MPC formulations, cost-function structures, and application domains. In addition, a cross-dimensional synthesis is conducted to identify recurring design patterns and reported associations among these dimensions within the reviewed corpus. The review highlights methodological trends, commonly adopted integration strategies, and recurring practical challenges, including computational burden, sample efficiency, robustness, and closed-loop guarantees. The resulting synthesis provides a structured reference for researchers and practitioners seeking to design or analyze RL--MPC architectures based on linear or linearized predictive control formulations.
Integrated deep reinforcement learning (DRL) and model predictive control (MPC) methods are increasingly used to control autonomous systems by combining their complementary capabilities. DRL learns control policies through interaction with the environment. MPC uses a system model to optimize control inputs while accounting for constraints. In DRL-MPC frameworks with shared control authority, both the DRL agent and the MPC controller each determine part of the control inputs. However, common learning formulations treat MPC as part of the environment and therefore do not explicitly account for MPC's contribution to control or its interaction with the DRL agent. This paper proposes a novel composite-gradient learning (CGL) method that integrates the MPC controller into the learning process by representing the DRL and MPC control inputs as a joint action and accounting for their interaction when updating the DRL agent during training. CGL is evaluated on two multi-class freeway traffic networks with different strengths of interaction between the DRL and MPC control inputs and it is compared with alternative methods that treat MPC as part of the environment or that only partially incorporate MPC into learning. The results show that CGL offers limited benefit under weak interaction, but learns higher-performing control policies than the alternative methods in a subset of training runs under strong interaction, although the average control-performance gains remain modest.
Sampling-based model predictive control methods handle nonlinear dynamics and complex cost landscapes through Monte Carlo rollouts, yet typically employ fixed constraint penalties that do not adapt to model-plant mismatch. This paper proposes Residual-Conservative Model Predictive Path Integral Control (RC-MPPI), a sampling-based MPC framework that modulates safety conservatism online using the prediction-execution residual. RC-MPPI combines three coupled mechanisms: residual-dependent constraint tightening, adaptive safety-cost shaping, and residual-adaptive sampling modulation through exploration contraction and temperature relaxation. The temperature adaptation reflects a key insight: when the model is inaccurate, rollout cost evaluations become unreliable, and increasing temperature reduces overcommitment to apparent cost rankings. Under Lipschitz dynamics and sub-Gaussian disturbances, we derive probabilistic bounds on constraint violation and show that the joint effect of the adaptive mechanisms reduces violation probability as the residual grows. A rollout-cost uncertainty analysis further shows that model-plant mismatch perturbs MPPI importance weights in proportion to residual magnitude and inversely with temperature, providing theoretical justification for residual-adaptive temperature relaxation. Simulations on an LTI point-mass system and a planar 2R manipulator show improved safety margin, success rate, and control efficiency compared with vanilla MPPI under significant model-plant mismatch.
Model Predictive Control (MPC) is widely used in industrial and robotic systems for enforcing constraints and embedding domain knowledge through finite-horizon optimization-based planning. However, despite these strengths, an MPC scheme typically does not yield optimal policies for sequential decision-making problems formulated as Markov Decision Processes (MDPs). Recent combinations of MPC with Reinforcement Learning (RL) alleviate this issue by treating MPC as a parameterized model of the optimal policy of an MDP and adjusting its parameters using data. While these approaches typically consider classical MDPs, many real-world problems include future information--such as forecasts, prices, or reference trajectories--at decision time, which must be included in the MDP state for optimal decision-making. Current MPC-RL approaches do not directly account for this augmented-state structure, raising the question of how to incorporate future information into MPC to obtain an optimal policy. This work establishes the structural requirements under which a parameterized MPC can exactly represent the optimal value functions and policy of an MDP with future information. We further demonstrate that such a parameterized MPC can serve as a structured function approximator, with its parameters learned using RL. The approach is illustrated on a point-mass racing task with future reference information.
Shambhuraj Sawant, Akhil S Anand, Dirk Reinhardt +1