Path planning method for computing optimal parking maneuvers for road vehicles and corresponding system

Patent metadata

Field	Value
Publication number	US-10960876-B2
Application number	US-201816021357-A
Country	US
Kind code	B2
Filing date	Jun 28, 2018
Priority date	Jun 30, 2017
Publication date	Mar 30, 2021
Grant date	Mar 30, 2021

Abstract

Path planning method for computing optimal parking maneuvers for road vehicles including the steps of computing a set of value functions of a cost function of parking maneuvers reaching the target set of states as unique viscosity solution of a Hamilton Jacobi Bellman equation, supplying the set of value functions, together with a starting state of the vehicle, as input to the dynamic programming calculation procedure calculating at least the set of vehicle controls. The set of equations modeling the evolution of the state of said road vehicle is a switched system of equations between a first sub-system if the vehicle is in forward motion and a second sub-system if the vehicle is in reverse motion. The cost function takes into account the arrival time a number of direction changes of the road vehicle between forward motion and reverse motion.

First claim

What is claimed is:

1. A path planning method for computing optimal parking maneuvers for road vehicles, in particular operating in a known environment in the presence of static obstacles, said method comprising operating according to a dynamic programming calculation procedure to compute a set of vehicle controls implementing an optimal parking maneuver to reach a target set of states corresponding to a given parking target, said parking maneuvers being trajectories obtained by a system of equations modeling the evolution of a state of said road vehicle as a function of vehicle controls,
said method including the steps of:
computing a set of value functions of a cost function of parking maneuvers reaching said target set of states as unique viscosity solution of a Hamilton-Jacobi-Bellman equation, said cost function taking in account an arrival time of the vehicle for a given parking maneuver,
supplying said set of value functions, together with a starting state of the vehicle, as input to a control synthesis procedure calculating at least said set of vehicle controls,
wherein said set of equations modeling the evolution of the state of said road vehicle is a switched system of equations switching between a first sub-system if the vehicle is in forward motion and a second sub-system if the vehicle is in reverse or backwards motion,
said cost function taking into account further to said arrival time a number of direction changes of the road vehicle between forward motion and reverse motion, said set of value functions of said cost function comprising a sequence of said value functions according to an index corresponding to said number of direction changes ranging from a zero number of direction changes to a maximum number of direction changes, and
wherein said cost function takes into account the number of direction changes between forward motion and reverse motion multiplied by a settable penalty parameter.

2. The method as set forth in claim 1 , wherein said set of vehicle controls includes a vector of controls and a vector of gear switchings.

3. The method as set forth in claim 1 , further including the step of defining an operating space of the vehicle, preferably partitioned as free space is the free space and space with obstacles.

4. The method as set forth in claim 1 , further including the step of rescaling of the value function as a rescaled value function.

5. The method as set forth in claim 1 , wherein said step of computing a set of value functions of a cost function of parking maneuvers reaching said target set of states as unique viscosity solution of a Hamilton-Jacobi-Bellman equation further includes the step of performing a numerical solving procedure which includes obtaining a discrete time Hamilton Jacobi Bellman equation, performing a discretization of the controls, computing a grid on a finite set of vertices in said op operating space, evaluating said discrete time Hamilton-Jacobi-Bellman equation on said set of vertices.

6. The method as set forth in claim 1 , wherein said dynamic programming procedure includes performing synthesis of feedback controls by a reiterated Euler integration step until the distance between the current and target state is less than an allowed distance tolerance.

7. The method as set forth in claim 1 , wherein said switched system is defined by a dynamics function which is a function of the state of the road vehicle, of a gear switching index, and of an angular velocity of the vehicle, a linear velocity being kept constant at a determined value.

8. A system for computing optimal parking maneuvers for road vehicles comprising at least a microprocessor based module on the vehicle which receives from vehicle modules information regarding the position and dynamic of the vehicle, wherein said system performs the path planning method for computing optimal parking maneuvers of claim 1 .

9. The system as set forth in claim 8 wherein said system supplies said set of vehicle controls to an automatic parking system.

10. The system as set forth in claim 8 wherein said system supplies said set of vehicle controls to an automatic parking system as assistance information to a human driver.

Assignees

• Magneti Marelli Spa
• Univ Degli Studi Di Parma

Inventors

• Micelli Piero
• Consolini Luca
• Laurini Mattia
• Locatelli Marco
• Furlan Axel
• Giorelli Michele

Classifications

• B62D15/027 Primary

  Parking aids, e.g. instruction means · CPC title

• B62D15/0285

  Parking performed automatically · CPC title

• G08G1/096833

  where different aspects are considered when computing the route · CPC title

• B60W50/14

  Means for informing the driver, warning the driver or prompting a driver intervention · CPC title

• B60W2050/0034

  Multiple-track, 2D vehicle model, e.g. four-wheel model · CPC title