A Receding-Horizon Framework for Co-Optimizing the Velocity and Power-Split of Automated Plug-In Hybrid Electric Vehicles

2021 ◽  
Vol 1 (4) ◽  
Author(s):  
Di Chen ◽  
Mike Huang ◽  
Anna Stefanopoulou ◽  
Youngki Kim
Keyword(s):  
Computation Time ◽  
Global Optimality ◽  
Receding Horizon ◽  
Prediction Horizon ◽  
Control Framework ◽  
Power Split ◽  
Online Implementation ◽  
Driving Scenario ◽  
Predictive Controller ◽  
Fuel Economy Improvement

Abstract This paper presents a control framework to co-optimize the velocity and power-split operation of a plug-in hybrid vehicle (PHEV) online in the presence of traffic constraints. The principal challenge in its online implementation lies in the conflict between the long control horizon required for global optimality and limits in available computational power. To resolve the conflict between the length of horizon and its computation complexity, we propose a receding-horizon strategy where co-states are used to approximate the future cost, helping to shorten the prediction horizon. In particular, we update the co-state using a nominal trajectory and the temporal-difference (TD) error based on co-state dynamics. Our simulation results demonstrate a 12% fuel economy improvement over the sequential/layered control strategy for a given driving scenario. Moreover, its real-time practicality is evidenced by a computation time per model predictive controller (MPC) step on average of around 80 ms within a 10 s prediction horizon.

scholarly journals A multivariable multiobjective predictive controller

2013 ◽  
Vol 23 (1) ◽  
pp. 35-45 ◽  
Author(s):  
Faten Ben Aicha ◽  
Faouzi Bouani ◽  
Mekki Ksouri
Keyword(s):  
Genetic Algorithm ◽  
Predictive Control ◽  
Mimo System ◽  
Weighting Factor ◽  
Nsga Ii ◽  
Weighted Sum Method ◽  
Prediction Horizon ◽  
Predictive Controller ◽  
Pareto Method ◽  
The Cost

Predictive control of MIMO processes is a challenging problem which requires the specification of a large number of tuning parameters (the prediction horizon, the control horizon and the cost weighting factor). In this context, the present paper compares two strategies to design a supervisor of the Multivariable Generalized Predictive Controller (MGPC), based on multiobjective optimization. Thus, the purpose of this work is the automatic adjustment of the MGPC synthesis by simultaneously minimizing a set of closed loop performances (the overshoot and the settling time for each output of the MIMO system). First, we adopt the Weighted Sum Method (WSM), which is an aggregative method combined with a Genetic Algorithm (GA) used to minimize a single criterion generated by the WSM. Second, we use the Non- Dominated Sorting Genetic Algorithm II (NSGA-II) as a Pareto method and we compare the results of both the methods. The performance of the two strategies in the adjustment of multivariable predictive control is illustrated by a simulation example. The simulation results confirm that a multiobjective, Pareto-based GA search yields a better performance than a single objective GA.

Effect of Short-Term Weather Predictions on Model Predictive Trajectory Tracking Performance of Unmanned Surface Vessels

2020 ◽  
Author(s):  
Benjamin Armentor ◽  
Joseph Stevens ◽  
Nathan Madsen ◽  
Andrew Durand ◽  
Joshua Vaughan
Keyword(s):  
Trajectory Tracking ◽  
Wind Waves ◽  
Tracking Error ◽  
Tracking Performance ◽  
Model Predictive Controller ◽  
Prediction Horizon ◽  
Surface Vessels ◽  
Waves And Currents ◽  
Predictive Controller ◽  
Disturbance Model

Abstract For mobile robots, such as Autonomous Surface Vessels (ASVs), limiting error from a target trajectory is necessary for effective and safe operation. This can be difficult when subjected to environmental disturbances like wind, waves, and currents. This work compares the tracking performance of an ASV using a Model Predictive Controller that includes a model of these disturbances. Two disturbance models are compared. One prediction model assumes the current disturbance measurements are constant over the entire prediction horizon. The other uses a statistical model of the disturbances over the prediction horizon. The Model Predictive Controller performance is also compared to a PI-controlled system under the same disturbance conditions. Including a disturbance model in the prediction of the dynamics decreases the trajectory tracking error over the entire disturbance spectrum, especially for longer horizon lengths.

scholarly journals Chance-Constrained Optimal Path Planning With Obstacles

2011 ◽  
Vol 27 (6) ◽  
pp. 1080-1094 ◽  
Author(s):  
Lars Blackmore ◽  
Masahiro Ono ◽  
Brian C. Williams
Keyword(s):  
Autonomous Vehicles ◽  
Solution Method ◽  
Optimal Path ◽  
Computation Time ◽  
Global Optimality ◽  
New Approach ◽  
Optimal Path Planning ◽  
Modeling Errors ◽  
Chance Constrained Approach ◽  
Probability Of Collision

Autonomous vehicles need to plan trajectories to a specified goal that avoid obstacles. For robust execution, we must take into account uncertainty, which arises due to uncertain localization, modeling errors, and disturbances. Prior work handled the case of set-bounded uncertainty. We present here a chance-constrained approach, which uses instead a probabilistic representation of uncertainty. The new approach plans the future probabilistic distribution of the vehicle state so that the probability of failure is below a specified threshold. Failure occurs when the vehicle collides with an obstacle or leaves an operator-specified region. The key idea behind the approach is to use bounds on the probability of collision to show that, for linear-Gaussian systems, we can approximate the nonconvex chance-constrained optimization problem as a disjunctive convex program. This can be solved to global optimality using branch-and-bound techniques. In order to improve computation time, we introduce a customized solution method that returns almost-optimal solutions along with a hard bound on the level of suboptimality. We present an empirical validation with an aircraft obstacle avoidance example.

Hybrid Powertrain Optimization With Real-Time Traffic Information

2013 ◽  
Author(s):  
Mohd Azrin Mohd Zulkefli ◽  
Jianfeng Zheng ◽  
Zongxuan Sun ◽  
Henry Liu
Keyword(s):  
Real Time ◽  
Computation Time ◽  
Lookup Table ◽  
Traffic Information ◽  
Hybrid Powertrain ◽  
Real Time Traffic ◽  
Necessary Conditions For Optimality ◽  
Online Implementation ◽  
Fuel Rate ◽  
Computation Load

Combining hybrid powertrain optimization with traffic information has been researched before, but tradeoffs between optimality, driving-cycle sensitivity and speed of calculation have not been cohesively addressed. Optimizing hybrid powertrain with traffic can be done through iterative methods such as Dynamic Programming (DP), Stochastic-DP and Model Predictive Control, but high computation load limits their online implementation. Equivalent Consumption Minimization Strategy (ECMS) and Adaptive-ECMS were proposed to minimize computation time, but unable to ensure real-time charge-sustaining-operation (CS) in transient traffic environment. Others show relationship between Pontryagin’s Minimum Principles (PMP) and ECMS, but iteratively solve the CS-operation problem offline. This paper proposes combining PMP’s necessary conditions for optimality, with sum-of State-Of-Charge-derivative for CS-operation. A lookup table is generated offline to interpolate linear mass-fuel-rate vs net-power-to-battery slopes to calculate the equivalence ratio for real-time implementation with predicted traffic data. Maximum fuel economy improvements of 7.2% over Rule-Based is achieved within a simulated traffic network.

scholarly journals Stabilize Nonlinear Interactive Process by Predictive Method on Receding Horizon

2019 ◽  
Vol 8 (10) ◽  
pp. 999-1004
Keyword(s):  
Point Estimate ◽  
Receding Horizon ◽  
Prediction Horizon ◽  
Nonlinear Predictive Control ◽  
Advance Method ◽  
Extended Interaction ◽  
Tuning Methods ◽  
And Control ◽  
Unstable Position ◽  
Sample Time

In this concept, analyze two processes model with certain tuning conditions. Design process with nonlinear conditions, disturbance and unstable position along with mutually interaction. Most of the processes are dynamic only w.r.t manipulated variable, disturbances and other unexpected conditions. Two models are resembled as interactive and mutual controlled process through sensors. Compare both processes with and without nonlinearity along with disturbance with tuning condition upto reach set point. Estimate prediction horizon, control horizon, weighted input, weighted rate of input, weighted output and desire trajectory are imposed for both processes by nonlinear predictive control. The main intense to regulate pressure and get desire level with impact of disturbances as temperature, added impurities etc… In this work, can be extended interaction of process with NMPC by adding more complexity and stabilize with certain tuning methods. NMPC is advance method to predictive input as well as output with respected sample time based on prediction horizon and control horizon on receding horizon of axis.

scholarly journals Model predictive Controller for Mobile Robot

2017 ◽  
Vol 2 (2) ◽  
pp. 18 ◽  
Author(s):  
Alireza Rezaee
Keyword(s):  
Mobile Robot ◽  
Adaptive Controller ◽  
Quadratic Equation ◽  
Linear Quadratic ◽  
Model Predictive Controller ◽  
Prediction Horizon ◽  
Future Behavior ◽  
Ricatti Equation ◽  
Predictive Controller ◽  
Real Robot

This paper proposes a Model Predictive Controller (MPC) for control of a P2AT mobile robot. MPC refers to a group of controllers that employ a distinctly identical model of process to predict its future behavior over an extended prediction horizon. The design of a MPC is formulated as an optimal control problem. Then this problem is considered as linear quadratic equation (LQR) and is solved by making use of Ricatti equation. To show the effectiveness of the proposed method this controller is implemented on a real robot. The comparison between a PID controller, adaptive controller, and the MPC illustrates advantage of the designed controller and its ability for exact control of the robot on a specified guide path.

An energy-time optimal autonomous motion control framework for overhead cranes in the presence of obstacles

2020 ◽  
pp. 095440622095450
Author(s):  
Xinwei Wang ◽  
Jie Liu ◽  
Xianzhou Dong ◽  
Haijun Peng ◽  
Chongwei Li
Keyword(s):  
Motion Control ◽  
Trajectory Tracking ◽  
Overhead Cranes ◽  
Control Framework ◽  
Weighted Matrix ◽  
Autonomous Motion ◽  
Avoidance Condition ◽  
Time Optimal ◽  
Predictive Controller ◽  
Time And Energy

This paper focuses on the autonomous motion control of 3-D underactuated overhead cranes in the presence of obstacles, and an “offline motion planning + online trajectory tracking” framework is developed. In the motion planner, to meet the balance between transfer time and energy consumption, the transfer mission is formulated as an energy-time hybrid optimal control problem. And a simple and conservative collision-avoidance condition is derived. To achieve fast and robust calculations, an iterative procedure that determines optimal terminal time based on the secant method is developed. Finally, to realize the high-precision trajectory tracking and fast residual sway suppression, a model predictive controller with a piecewise weighted matrix is designed. Numerical simulation demonstrates that the discussed framework is effective.

Robust overtaking control of autonomous electric vehicle with parameter uncertainties

2019 ◽  
Vol 233 (13) ◽  
pp. 3358-3376 ◽  
Author(s):  
Liwei Xu ◽  
Weichao Zhuang ◽  
Guodong Yin ◽  
Guangmin Li ◽  
Chentong Bian
Keyword(s):  
Tracking Control ◽  
Lower Layer ◽  
Vehicle Stability ◽  
Reference Trajectory ◽  
Parameter Uncertainties ◽  
Ground Vehicle ◽  
Robust Model ◽  
Control Framework ◽  
Predictive Controller ◽  
State Tracking

In this paper, we propose a two-layer overtaking control framework of the autonomous electric ground vehicle, including trajectory planning and tracking control. In the upper layer, a stable overtaking trajectory is derived by considering the duration of the overtaking, driving comfort and vehicle stability simultaneously. Besides, the safety spacing between the overtaking and overtaken vehicles is analyzed to prevent the potential collision. To track the generated reference trajectory precisely, we design a robust model predictive controller adopting the input-to-state stability property in the lower layer to reduce the affection of disturbances and parameter uncertainties. Two driving strategies with different lateral accelerations, that is, mild and aggressive, are simulated to verify the effectiveness of the proposed control framework. By comparing to the driver-vehicle control system, the proposed control framework can not only achieve safe, smooth, and rapid overtaking but also realize the accurate state tracking in the presence of tire cornering stiffness uncertainty.

Sign in / Sign up

Export Citation Format

Share Document