PSO

python_motion_planning.global_planner.evolutionary_search.pso.PSO

Bases: EvolutionarySearcher

Class for Particle Swarm Optimization (PSO) motion planning.

Parameters:

Name	Type	Description	Default
start	tuple	start point coordinate	required
goal	tuple	goal point coordinate	required
env	Env	environment	required
heuristic_type	str	heuristic function type	'euclidean'
n_particles	int	number of particles	300
w_inertial	float	inertial weight	1.0
w_cognitive	float	cognitive weight	1.0
w_social	float	social weight	1.0
point_num	int	number of position points contained in each particle	5
max_speed	int	The maximum velocity of particle motion	6
max_iter	int	maximum iterations	200
init_mode	int	Set the generation mode for the initial position points of the particle swarm	GEN_MODE_RANDOM

Examples:

Python Console Session

>>> import python_motion_planning as pmp
>>> planner = pmp.PSO((5, 5), (45, 25), pmp.Grid(51, 31))
>>> cost, path, fitness_history = planner.plan(verbose=True)     # planning results only
>>> cost_curve = [-f for f in fitness_history]
>>> planner.plot.animation(path, str(planner), cost, cost_curve=cost_curve)  # animation
>>> planner.run()       # run both planning and animation

calFitnessValue(position)

Calculate the value of fitness function.

Parameters:

Name	Type	Description	Default
position	list	control points calculated by PSO	required

Returns:

Name	Type	Description
fitness	float	the value of fitness function

initializePositions()

Generate n particles with pointNum_ positions each within the map range.

Returns:

Name	Type	Description
init_positions	list	initial position sequence of particle swarm

isCollision(p1, p2)

Judge collision when moving from node1 to node2 using Bresenham.

Parameters:

Name	Type	Description	Default
p1	tuple	start point	required
p2	tuple	end point	required

Returns:

Name	Type	Description
collision	bool	True if collision exists, False otherwise.

optimizeParticle(particle)

Particle update optimization iteration

Parameters:

Name	Type	Description	Default
particle	Particle	the particle	required

plan(verbose=False)

Particle Swarm Optimization (PSO) motion plan function.

Parameters:

Name	Type	Description	Default
verbose	bool	print the best fitness value of each iteration	False

Returns:

Name	Type	Description
cost	float	path cost
path	list	planning path

run()

Running both plannig and animation.

updateParticlePosition(particle)

A function to update the particle position

Parameters:

Name	Type	Description	Default
particle	Particle	the particle	required

updateParticleVelocity(particle)

A function to update the particle velocity

Parameters:

Name	Type	Description	Default
particle	Particle	the particle	required