Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map

Buyval, Alexander; Gabdullin, Aidar; Klimchik, Alexandr

doi:10.5220/0006901702480254

Cited by 4 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The original version of the MPPI algorithm [6] considers an offline static cost map with previous knowledge of the track configuration. It has been extended to environments with dynamic obstacles in [9]. Also, a study on partially observable scenarios for Quadrotors was introduced in [10].…”

Section: Introductionmentioning

confidence: 99%

Minimal Injury Risk Motion Planning Using Active Mitigation and Sampling Model Predictive Control

Guardini

Spalanzani

Martinet

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

Collision mitigation is an important element in motion planning. Although Advanced Driver-Assistance Systems (ADAS) have a rich number of functionalities, they lack interchangeability. There is still a gap on finding a way to evaluate the best decision globally. This paper presents a novel motion planning framework to generate emergency maneuvers in complex and risky scenarios using active mitigation. The classical Model Predictive Path Integral (MPPI) algorithm is improved to be used in a probabilistic dynamic cost map under limited perception range. A cost map with global probability of injury to all road users is used as a constraint to the problem in order to compute target selection based on the global minimum risk considering all road users. Real experiments introduce the use of augmented sensor data by merging simulation and real sensor data to safely produce collision and mitigation experiments. Results show that the proposed algorithm can perform correctly in real time on board of the vehicle, by finding collision-free trajectories in complex scenarios and compute viable target selection that minimizes global injury risk when collision is inevitable.

show abstract

Section: Introductionmentioning

confidence: 99%

Minimal Injury Risk Motion Planning Using Active Mitigation and Sampling Model Predictive Control

Guardini

Spalanzani

Martinet

et al. 2022

2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

show abstract

“…Two key contribution steps of this research/paper stand out: The first step is: Test the vehicle model with 4-wheel drive in the computational form [16][17], this being the objective of the work. The second step (for later work): Validate it on a platform/simulator (Hardware in the Loop).…”

Section: Introductionmentioning

confidence: 99%

Longitudinal Modeling of a Road Vehicle: 4-Wheel Traction

Manuel

Santos

Lenzi

et al. 2022

IJRCS

View full text Add to dashboard Cite

This paper presents the longitudinal modeling of a 4-wheel traction vehicle represented in a block diagram using Matlab®/Simulink® software. The proposed modeling is suitable to be implemented in automatic parallel, oblique, or perpendicular parking systems considering speed cases between 5 km/h and 30 km/h. For the computational simulations, it was considered that the vehicle starts at rest and goes up a referenced or determined slope in degrees (°), with a sufficient rear reaction force to allow the vehicle to move until the engine produces sufficient torque. For the model of the tire variant, the magic formula (characterized by the sum of five vectors about an axis) was used. Three input signals were considered, slope, wind, and accelerator variation were considered in numerical simulations. The output signals are rear and normal front forces, vehicle speed, angular velocity, and engine acceleration. The longitudinal modeling proposed allows for easily reproducing the results and assigning new parameters to validate a Project, contributing positively both to the automotive industries and in innovation-based scientific research.

show abstract

Motion control of self-driving vehicles on rough roads

Shapovalov

Gafurov

Klimchik

2019

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Control of self-driving vehicles is one of the-up-to-date tasks. Vehicles are non-linear dynamic systems with a large number of parameters. When it is supposed to move on a rough road with a lot of abrupt turns, significant unpredictable and unmeasurable disturbances emerge and the control quality degrades significantly. We implemented model predictive path integral (MPPI) control to deal with significant unmeasured disturbances. To approve efficiency of the proposed controller, we conducted simulation of complex car motion on a plane under the control of PID, and MPPI controllers. Performance of controllers is comparatively equal in stable modes, when reference position and velocity do not change. MPPI controllers outperform PID ones in unstable modes due to their ability to predict a vehicle behavior. Also, MPPI is the best choice for the rejection of significant external disturbances.

show abstract

Model Predictive Path Integral Control for Car Driving with Dynamic Cost Map

Cited by 4 publications

References 4 publications

Minimal Injury Risk Motion Planning Using Active Mitigation and Sampling Model Predictive Control

Minimal Injury Risk Motion Planning Using Active Mitigation and Sampling Model Predictive Control

Longitudinal Modeling of a Road Vehicle: 4-Wheel Traction

Motion control of self-driving vehicles on rough roads

Contact Info

Product

Resources

About