Abstract:We present a chance-constrained model predictive control (MPC) framework under Gaussian mixture model (GMM) uncertainty. Specifically, we consider the uncertainty that arises from predicting future behaviors of moving obstacles, which may exhibit multiple modes (for example, turning left or right). To address the multi-modal uncertainty distribution, we propose three MPC formulations: nominal chance-constrained planning, robust chance-constrained planning, and contingency planning. We prove that closed-loop trajectories generated by the three planners are safe. The approaches differ in conservativeness and performance guarantee. In particular, the robust chance-constrained planner is recursively feasible under certain assumptions on the propagation of prediction uncertainty. On the other hand, the contingency planner generates a less conservative closed-loop trajectory than the nominal planner. We validate our planners using state-of-the-art trajectory prediction algorithms in autonomous driving simulators.

What problem does this paper attempt to address?

### Problems Addressed by the Paper The paper aims to address the problem of safe trajectory planning for autonomous driving vehicles under multimodal uncertainty distributions (such as Gaussian Mixture Models, GMM). Specifically, the researchers focus on how to plan collision-free trajectories in the presence of moving obstacles (such as other vehicles or pedestrians) whose future movements may have multiple modes (e.g., going straight or turning). The paper proposes three Model Predictive Control (MPC) frameworks: nominal chance-constrained planning, robust chance-constrained planning, and emergency planning, and demonstrates that the closed-loop trajectories generated by these three methods are safe. These methods differ in terms of conservativeness and performance guarantees, with robust chance-constrained planning ensuring recursive feasibility under certain assumptions, while emergency planning produces less conservative closed-loop trajectories compared to nominal planning. ### Main Contributions 1. **Recursively Feasible and Safe MPC Framework**: To achieve this, the authors formulate conditions on the propagation of GMM during prediction and accordingly tighten the MPC constraints to counteract the uncertainties brought by prediction propagation. 2. **Emergency Planning-Based Alternative**: An emergency planning-based method is proposed to reduce the conservativeness of chance-constrained MPC under GMM uncertainty while maintaining safety guarantees. 3. **Simulation Validation**: The effectiveness of the methods is validated through simulations using the state-of-the-art trajectory prediction model Trajectron++ and the autonomous driving simulator CARLA. ### Method Comparison By comparing different risk-constrained methods (including methods based on exact GMM information, data-driven methods, etc.), the study finds that when the uncertainty conforms to GMM, using chance constraints as a risk measure combined with distribution information (such as M1 and M3) can provide less conservative results while ensuring computational efficiency. In contrast, using CVaR constraints or distributional robustness may lead to overly conservative results and significantly increase computation time. ### Robust Planning to Ensure Recursive Feasibility To ensure the recursive feasibility of shrinking horizon MPC in interactive scenarios, the paper proposes a robust planning scheme and conservatively approximates this scheme based on the Cauchy-Schwarz inequality. In this way, the planning process can ensure recursive feasibility under certain assumptions. In summary, the paper is dedicated to developing a provably safe and computationally efficient trajectory planning method under multimodal uncertain environments, particularly suitable for autonomous driving applications in complex dynamic environments.

Safe Chance-constrained Model Predictive Control under Gaussian Mixture Model Uncertainty

Chance Constrained Nonlinear Model Predictive Control

Chance Constrained Model Predictive Control Via Active Uncertainty Set Learning and Calibration

Learning-based Model Predictive Control for Safe Path Planning and Control

Stochastic MPC for Finite Gaussian Mixture Disturbances with Guarantees

An Adaptive Model Predictive Control Strategy for a Class of Discrete‐time Linear Systems with Parametric Uncertainty

Moment-Based Model Predictive Control of Autonomous Systems

A Chance‐constrained Tube‐based Model Predictive Control for Tracking Linear Systems Using Data‐driven Uncertainty Sets

Moving Obstacle Collision Avoidance via Chance-Constrained MPC with CBF

Contingency Model Predictive Control for Linear Time-Varying Systems

Iterative risk-constrained model predictive control: A data-driven distributionally robust approach

Chance-Constrained Information-Theoretic Stochastic Model Predictive Control with Safety Shielding

Predictive Control for Autonomous Driving with Uncertain, Multi-modal Predictions

Correntropy based Model Predictive Controller with Multi-constraints for Robust Path Trajectory Tracking of Self-driving Vehicle

Chance-Constrained Convex MPC for Robust Quadruped Locomotion Under Parametric and Additive Uncertainties

Probabilistically safe controllers based on control barrier functions and scenario model predictive control

A Robust Mpc Approach for Platooning Control of Automated Vehicles with Constraints on Acceleration

Data-driven distributionally robust iterative risk-constrained model predictive control

Integrating Predictive Motion Uncertainties with Distributionally Robust Risk-Aware Control for Safe Robot Navigation in Crowds

Robust Learning-based Model Predictive Control for Intelligent Vehicles with Unknown Dynamics and Unbounded Disturbances

Backup Plan Constrained Model Predictive Control with Guaranteed Stability