I'm trying to create a simulation for a conveyor belt with objects on it, to see how the behaviour would be. I am playing around but struggling with the functionality of the conveyor belt. Right now, I try to add impulse or velocity to the objects whenever they have a collision with the belt, but I have yet to get a good result. Here is just an example of how I tried to give impulse to the objects, however, whenever I add a new object (with mouse event), the new impulse only applies to the newest object.
import pygame
import pymunk
import pymunk.pygame_util
import math
import sys
pygame.init()
space = pymunk.Space()
WIDTH, HEIGHT = 1920,800
mu = 2
window = pygame.display.set_mode((WIDTH, HEIGHT))
pygame.display.set_caption("Kinematics approximation ")
def calculate_distance(p1, p2):
return math.sqrt((p2[1] - p1[1])**2 + (p2[0] - p1[0])**2)
def calculate_angle(p1, p2):
return math.atan2(p2[1] - p1[1], p2[0] - p1[0])
def draw(space, window,draw_options):
window.fill("white")
space.debug_draw(draw_options)
pygame.display.update()
def create_belt(space):
belts = [
[(0,500), (600,500), 5],
[(600,500), (800,400), 6],
[(800,400), (WIDTH,400), 6]
]
for pos_1, pos_2, width in belts:
shape = pymunk.Segment(space.static_body,pos_1, pos_2, width)
shape.body.position = 0,0
shape.friction = mu
space.add(shape)
def create_object(space, mass, pos):
body = pymunk.Body(body_type = pymunk.Body.DYNAMIC)
body.position = pos
shape = pymunk.Poly.create_box(body, size = (20,10))
shape.mass = mass
shape.color = (255,0,0, 100)
shape.body.friction = mu
#shape.body.velocity = (80,0)
space.add(body,shape)
return shape
def run(window, width, height):
run = True
clock = pygame.time.Clock()
fps = 240
dt = 1/fps
space = pymunk.Space()
space.gravity = (0,981)
create_belt(space)
def coll_begin(arbiter, space, data):
print(arbiter)
#angle = calculate_angle(*)
#force = calculate_distance(*line) * 50
#fx = math.cos(angle) * force
#fy = math.sin(angle) * force
return True
def coll_pre(arbiter, space, data):
for object in objects:
object.body.apply_impulse_at_local_point((12, 0),(0,0))
return True
def coll_post(arbiter, space, data):
#print(velocity_at_local_point)
pass
def coll_separate(arbiter, space, data):
pass
handler = space.add_default_collision_handler()
handler.begin = coll_begin
handler.pre_solve = coll_pre
handler.post_solve = coll_post
handler.separate = coll_separate
draw_options = pymunk.pygame_util.DrawOptions(window)
pressed_pos = None
while run:
for event in pygame.event.get():
if event.type == pygame.QUIT:
pygame.quit()
sys.exit()
if event.type == pygame.MOUSEBUTTONDOWN:
pressed_pos = pygame.mouse.get_pos()
objects = []
objects.append(create_object(space,10, pressed_pos))
draw(space,window, draw_options)
space.step(dt)
clock.tick(fps)
pygame.quit()
sys.exit()
if __name__ == "__main__":
run(window, WIDTH, HEIGHT)
There are two problems in your code:
First, you are resetting the objects list every time a new object is created. This means that when you apply impulse to all objects in the objects list, only impulse is applied to the last object created. This can be easily fixed by moving the objects = [] line out of the game loop.
After fixing this, you might have noticed that all objects are getting impulse, whether they touch the belt or not. This is caused by the following: whenever any object is colliding with the conveyor belt, you apply an impulse to all objects. This causes all objects to move forward, not only the one that collides with the conveyor belt.
You can fix the two above errors using the Arbiter.shapes tuple that is provided as an argument to the coll_pre function. This tuple contains the two objects colliding with eachother, and thus also the object you want to move. The second object is the new object you want to move. This change also makes the objects list useless, as you don't use it anymore to apply the impulse.
Here's the code:
def coll_pre(arbiter, space, data):
print(arbiter)
arbiter.shapes[1].body.apply_impulse_at_local_point((12, 0),(0,0))
return True
I would also recommend increasing the impulse given, as now the objects travel to slowly to resist to gravity on the inclined part of the belt. A value of 40 worked for me.
Related
I am an aerospace student working on a school project for our python programming course. The assignment is create a program only using Pygame and numpy. I decided to create a wind tunnel simulation that simulates the airflow over a two dimensional wing. I was wondering if there is a more efficient way of doing the computation from a programming perspective. I will explain the program:
I have attached an image here:
The (steady) flow field is modeled using the vortex panel method. Basically, I am using a grid of Nx times Ny points where at each point a velocity (u,v) vector is given. Then using Pygame I map these grid points as circles, so that they resemble an area of influence. The grid points are the grey circles in the following image:
I create N particles and determine their velocities by iterating as follows:
create a list of particles.
create a grid list.
for each gridpoint in grid list:
for each particle in list of particles:
if particle A is within the area of influence of grid point n (xn,yn): particle A its velocity = velocity at grid point n.
Visualize everything in Pygame.
this basic way was the only way I could think of visualizing the flow in Pygame. The simulation works pretty well, but If I increase the number of grid points(increase the accuracy of the flow field), the performance decreases. My question is if there is a more efficient way to do this just using pygame and numpy?
I have attached the code here:
import pygame,random,sys,numpy
from Flow import Compute
from pygame.locals import *
import random, math, sys
#from PIL import Image
pygame.init()
Surface = pygame.display.set_mode((1000,600))
#read the airfoil geometry from a dat file
with open ('./resources/naca0012.dat') as file_name:
x, y = numpy.loadtxt(file_name, dtype=float, delimiter='\t', unpack=True)
#parameters used to describe the flow
Nx=30# 30 column grid
Ny=10#10 row grid
N=20#number of panels
alpha=0#angle of attack
u_inf=1#freestream velocity
#compute the flow field
u,v,X,Y= Compute(x,y,N,alpha,u_inf,Nx,Ny)
#The lists used for iteration
Circles = []
Particles= []
Velocities=[]
#Scaling factors used to properly map the potential flow datapoints into Pygame
magnitude=400
vmag=30
umag=30
panel_x= numpy.multiply(x,magnitude)+315
panel_y= numpy.multiply(-y,magnitude)+308
#build the grid suited for Pygame
grid_x= numpy.multiply(X,magnitude)+300
grid_y= numpy.multiply(Y,-1*magnitude)+300
grid_u =numpy.multiply(u,umag)
grid_v =numpy.multiply(v,-vmag)
panelcoordinates= zip(panel_x, panel_y)
# a grid area
class Circle:
def __init__(self,xpos,ypos,vx,vy):
self.radius=16
self.x = xpos
self.y = ypos
self.speedx = 0
self.speedy = 0
#create the grid list
for i in range(Ny):
for s in range(Nx):
Circles.append(Circle(int(grid_x[i][s]),int(grid_y[i][s]),grid_u[i][s],grid_v[i][s]))
Velocities.append((grid_u[i][s],grid_v[i][s]))
#a particle
class Particle:
def __init__(self,xpos,ypos,vx,vy):
self.image = pygame.Surface([10, 10])
self.image.fill((150,0,0))
self.rect = self.image.get_rect()
self.width=4
self.height=4
self.radius =2
self.x = xpos
self.y = ypos
self.speedx = 30
self.speedy = 0
#change particle velocity if collision with grid point
def CircleCollide(Circle,Particle):
Particle.speedx = int(Velocities[Circles.index((Circle))][0])
Particle.speedy = int(Velocities[Circles.index((Circle))][1])
#movement of particles
def Move():
for Particle in Particles:
Particle.x += Particle.speedx
Particle.y += Particle.speedy
#create particle streak
def Spawn(number_of_particles):
for i in range(number_of_particles):
i=i*(300/number_of_particles)
Particles.append(Particle(0, 160+i,1,0))
#create particles again if particles are out of wake
def Respawn(number_of_particles):
for Particle in Particles:
if Particle.x >1100:
Particles.remove(Particle)
if Particles==[]:
Spawn(number_of_particles)
#Collsion detection using pythagoras and distance formula
def CollisionDetect():
for Circle in Circles:
for Particle in Particles:
if Particle.y >430 or Particle.y<160:
Particles.remove(Particle)
if math.sqrt( ((Circle.x-Particle.x)**2) + ((Circle.y-Particle.y)**2) ) <= (Circle.radius+Particle.radius):
CircleCollide(Circle,Particle)
#draw everything
def Draw():
Surface.fill((255,255,255))
#Surface.blit(bg,(-300,-83))
for Circle in Circles:
pygame.draw.circle(Surface,(245,245,245),(Circle.x,Circle.y),Circle.radius)
for Particle in Particles:
pygame.draw.rect(Surface,(150,0,0),(Particle.x,Particle.y,Particle.width,Particle.height),0)
#pygame.draw.rect(Surface,(245,245,245),(Circle.x,Circle.y,1,16),0)
for i in range(len(panelcoordinates)-1):
pygame.draw.line(Surface,(0,0,0),panelcoordinates[i],panelcoordinates[i+1],3)
pygame.display.flip()
def GetInput():
keystate = pygame.key.get_pressed()
for event in pygame.event.get():
if event.type == QUIT or keystate[K_ESCAPE]:
pygame.quit();sys.exit()
def main():
#bg = pygame.image.load("pressure.png")
#bg = pygame.transform.scale(bg,(1600,800))
#thesize= bg.get_rect()
#bg= bg.convert()
number_of_particles=10
Spawn(number_of_particles)
clock = pygame.time.Clock()
while True:
ticks = clock.tick(60)
GetInput()
CollisionDetect()
Move()
Respawn(number_of_particles)
Draw()
if __name__ == '__main__': main()
The code requires another script that computes the flow field itself. It also reads datapoints from a textfile to get the geometry of the wing.
I have not provided these two files, but I can add them if necessary. Thank you in advance.
One bottleneck in your code is likely collision detection. CollisionDetect() computes the distance between each particle and each circle. Then, if a collision is detected, CircleCollide() finds the index of the circle in Circles (a linear search), so that the velocities can be retrieved from the same index in Velocities. Clearly this is ripe for improvement.
First, the Circle class already has the velocities in the speedx/speedy attributes, so Velocities can be eliminated .
Second, because the circles are at fixed locations, you can calculate which circle is closest to any given particle from the position of the particle.
# You may already have these values from creating grid_x etc.
# if not, you only need to calculated them once, because the
# circles don't move
circle_spacing_x = Circles[1].x - Circles[0].x
circle_spacing_y = Circles[Nx].y - Circles[0].y
circle_first_x = Circles[0].x - circle_spacing_x / 2
circle_first_y = Circles[0].y - circle_spacing_y / 2
Then CollisionDetect() becomes:
def CollisionDetect():
for particle in Particles:
if particle.y >430 or particle.y<160:
Particles.remove(particle)
continue
c = (particle.x - circle_first_x) // circle_spacing_x
r = (particle.y - circle_first_y) // circle_spacing_y
circle = Circles[r*Nx + c]
if ((circle.x - particle.x)**2 + (circle.y - particle.y)**2
<= (circle.radius+particle.radius)**2):
particle.speedx = int(circle.speedx)
particle.speedy = int(circle.speedy)
I've tidied up your code and made a few changes, namely adding scope to your classes and introducing a couple more. Without further knowledge of Flow I cannot test this fully, but if you could get back to me I can do some more. I'm assuming here that the 'flow field' can be simulated by the numpy.meshgrid function.
import pygame,numpy,sys
import pygame.locals
import math
class Particle:
def __init__(self,xpos,ypos,vx,vy):
self.size = numpy.array([4,4])
self.radius =2
self.pos = numpy.array([xpos,ypos])
self.speed = numpy.array([30,0])
self.rectangle = numpy.hstack((self.pos,self.size))
def move(self):
self.pos += self.speed
self.rectangle = numpy.hstack((self.pos,self.size))
def distance(self,circle1):
return math.sqrt(numpy.sum((circle1.pos - self.pos)**2))
def collision(self,circle1):
result = False
if self.pos[1] >430 or self.pos[1]<160:
result = True
if self.distance(circle1) <= (circle1.radius+self.radius):
self.speed = circle1.speed
return result
class Particles:
def __init__(self,num_particles):
self.num = num_particles
self.particles =[]
self.spawn()
def spawn(self):
for i in range(self.num):
i=i*(300/self.num)
self.particles.append(Particle(0, 160+i,1,0))
def move(self):
for particle in self.particles:
particle.move()
if particle.pos[0] >1100:
self.particles.remove(particle)
if not self.particles: self.spawn()
def draw(self):
for particle in self.particles:
pygame.draw.rect(Surface,(150,0,0),particle.rectangle,0)
def collisiondetect(self,circle1):
for particle in self.particles:
if particle.collision(circle1):
self.particles.remove(particle)
def GetInput():
keystate = pygame.key.get_pressed()
for event in pygame.event.get():
if event.type == pygame.locals.QUIT or keystate[pygame.locals.K_ESCAPE]:
pygame.quit()
sys.exit()
#draw everything
def Draw(sw,cir):
Surface.fill((255,255,255))
cir.draw()
for i in range(panelcoordinates.shape[1]):
pygame.draw.line(Surface,(0,0,0),panelcoordinates[0,i-1],panelcoordinates[0,i],3)
sw.draw()
pygame.display.flip()
# a grid area
class Circle:
def __init__(self,xpos,ypos,vx,vy):
self.radius=16
self.pos = numpy.array([xpos,ypos])
self.speed = numpy.array([vx,vy])
class Circles:
def __init__(self,columns,rows):
self.list = []
grid_x,grid_y = numpy.meshgrid(numpy.linspace(0,1000,columns),numpy.linspace(200,400,rows))
grid_u,grid_v = numpy.meshgrid(numpy.linspace(20,20,columns),numpy.linspace(-1,1,rows))
for y in range(rows):
for x in range(columns):
c1= Circle(int(grid_x[y,x]),int(grid_y[y,x]),grid_u[y,x],grid_v[y,x])
self.list.append(c1)
def draw(self):
for circle in self.list:
pygame.draw.circle(Surface,(245,245,245),circle.pos,circle.radius)
def detectcollision(self,parts):
for circle in self.list:
parts.collisiondetect(circle)
if __name__ == '__main__':
#initialise variables
number_of_particles=10
Nx=30
Ny=10
#circles and particles
circles1 = Circles(Nx,Ny)
particles1 = Particles(number_of_particles)
#read the airfoil geometry
panel_x = numpy.array([400,425,450,500,600,500,450,425,400])
panel_y = numpy.array([300,325,330,320,300,280,270,275,300])
panelcoordinates= numpy.dstack((panel_x,panel_y))
#initialise PyGame
pygame.init()
clock = pygame.time.Clock()
Surface = pygame.display.set_mode((1000,600))
while True:
ticks = clock.tick(6)
GetInput()
circles1.detectcollision(particles1)
particles1.move()
Draw(particles1,circles1)
I've also made some a crude stab at drawing a aerofoil, as again I don't have the knowledge of the data file with the coordinates.
I am creating tetris using pygame. i want to use collision detection so that when the shape in play comes into contact with any other previously played shapes, i can stop the shape, as per the logic of tetris. i came across pixel perfect collision using masks. i have followed some tutorials online, however the pixel detection returns true every time a new shape comes into play, not when any shapes collide. sorry in advance for the long code, its the bare minimum for the code to actually and still containing the game element of it. i think there is something wrong with my approach which is causing this error. I basically have a function that everytime the shape in play comes into contact with the 'floor' that shape is held in that position and a new shape is created. i think ive overcomplicated it, in turn creating this error. thanks in advance
import pygame
import sys
import shapelogic
pygame.init()
screensize = width, height = 800, 595
screen = pygame.display.set_mode(screensize)
background_image =pygame.image.load("/Users/marceason/PycharmProjects/Tetris/Wooden_background.jpg").convert_alpha()
myshape = 0
stop_movement = 0
blit_count = 0
stored_shapes = pygame.sprite.Group()
stored_shapes_with_coords = []
extra_blit_required = False
index = 0
count = 0
listofshapes = []
class shapemanager():
def __init__(self):
self.listofshapes = []
def create_another_instance(self):
global count
count += 1
string = "Shape_{0},".format(count)
another_shape = Shape(string)
self.listofshapes.append(another_shape)
global index
object = self.listofshapes[index]
index += 1
return object
def load_shape(self):
shape = self.create_another_instance()
shape.load_shapes()
class Shape(pygame.sprite.Sprite):
def __init__(self, name):
pygame.sprite.Sprite.__init__(self)
self.name = name
self.x = 50
self.y = 100
self.move_event = pygame.USEREVENT + 1
self.reached_bottom_event = pygame.USEREVENT + 2
self.one_sec_timer = 1000
self.half_sec_timer = 500
self.reachbottomflag = False
self.movement_possible = True
self.image = pygame.image.load(
"/Users/marceason/PycharmProjects/Tetris/Tetris_Shapes/Green_Shape_1_Position_1.png")
self.mask = pygame.mask.from_surface(self.image)
self.rect = self.image.get_rect()
def move_shape(self):
if self.movement_possible:
key_input = pygame.key.get_pressed()
if key_input[pygame.K_LEFT]:
self.x -= 16
if key_input[pygame.K_RIGHT]:
self.x += 16
if not self.reachbottomflag:
if key_input[pygame.K_DOWN]:
self.y += 16
def reachbottom(self):
if self.y >= 560:
self.reachbottomflag = True
def no_movement_possible(self):
self.movement_possible = False
def assign_shape():
global myshape
global stop_movement
myshape = sl.create_another_instance()
pygame.time.set_timer(myshape.move_event, myshape.one_sec_timer)
stop_movement = pygame.time.set_timer(myshape.reached_bottom_event, myshape.half_sec_timer)
def blit_used_shapes():
global screen
global blit_count
blit_count = len(stored_shapes_with_coords)
local_count = 0
while local_count < blit_count:
screen.blit(stored_shapes_with_coords[local_count][0], (stored_shapes_with_coords[local_count][1], stored_shapes_with_coords[local_count][2]))
local_count += 1
sl = shapemanager()
##### HERE IS THE PIXEL DETECTION #####
result = pygame.sprite.spritecollide(myshape, stored_shapes, False, pygame.sprite.collide_mask)
## Main loop ##
assign_shape()
while True:
for event in pygame.event.get():
if event.type == pygame.QUIT: sys.exit()
screen.blit(background_image, (0, 0))
screen.blit(myshape.image, (myshape.x, myshape.y))
myshape.move_shape()
key_input = pygame.key.get_pressed()
if key_input[pygame.K_SPACE]:
myshape.rotate_shape()
myshape.reachbottom()
if myshape.reachbottomflag:
if event.type == myshape.reached_bottom_event:
myshape.no_movement_possible()
stored_shape_tuple = [myshape.image, myshape.x, myshape.y]
stored_shapes_with_coords.append(stored_shape_tuple)
stored_shapes.add(myshape)
extra_blit_required = True
assign_shape()
####### PIXEL DETECTION IS HERE IN FOR LOOP ####
if result:
print("this should only execute when two shapes touch!!")
if extra_blit_required:
blit_used_shapes()
pygame.display.update()
The issue is that you are not updating the sprites rect attribute. The sprites rects all have position (0, 0) (since you do not set it in the call to self.image.get_rect()) and as a result the masks will all overlap and collide.
If you read the docs for pygame.sprite.collide_mask you will note that it says that your sprites need to have mask and rect attributes. You have a rect in your sprite and you set it in the __init__(), but you do not keep it updated when you move the sprite. You just change the x and y attributes without adjusting the rect position. The reason that the collide_mask wants a rect is that it uses that to determine the offset parameter for the pygame.mask.Mask.overlap() call that it uses. The important thing to realize is that masks themselves do not have a position, they need the rects to determine the relative positions of the masks.
This is similar to images/surfaces not having a position and needing a rect to track that for them.
On a separate issue, the way you are blit'ing the sprites to the screen makes no sense. You are not using the abilities of the sprite groups to draw and worse you are keeping the image, x and y of the sprite in a separate list and not containing it in the sprite itself. You should go look at some examples of pygame sprite based code. There are lots of examples out there.
Why is the physics wrong in the following Pymunk example?
from __future__ import print_function
import sys
from math import pi
import pygame
from pygame.locals import USEREVENT, QUIT, KEYDOWN, KEYUP, K_s, K_r, K_q, K_ESCAPE, K_UP, K_DOWN, K_LEFT, K_RIGHT
from pygame.color import THECOLORS
import pymunk
from pymunk import Vec2d
import pymunk.pygame_util
LEG_GROUP = 1
class Simulator(object):
def __init__(self):
self.display_flags = 0
self.display_size = (600, 600)
self.space = pymunk.Space()
self.space.gravity = (0.0, -1900.0)
self.space.damping = 0.999 # to prevent it from blowing up.
# Pymunk physics coordinates start from the lower right-hand corner of the screen.
self.ground_y = 100
ground = pymunk.Segment(self.space.static_body, (5, self.ground_y), (595, self.ground_y), 1.0)
ground.friction = 1.0
self.space.add(ground)
self.screen = None
self.draw_options = None
def reset_bodies(self):
for body in self.space.bodies:
if not hasattr(body, 'start_position'):
continue
body.position = Vec2d(body.start_position)
body.force = 0, 0
body.torque = 0
body.velocity = 0, 0
body.angular_velocity = 0
body.angle = body.start_angle
def draw(self):
### Clear the screen
self.screen.fill(THECOLORS["white"])
### Draw space
self.space.debug_draw(self.draw_options)
### All done, lets flip the display
pygame.display.flip()
def main(self):
pygame.init()
self.screen = pygame.display.set_mode(self.display_size, self.display_flags)
width, height = self.screen.get_size()
self.draw_options = pymunk.pygame_util.DrawOptions(self.screen)
def to_pygame(p):
"""Small hack to convert pymunk to pygame coordinates"""
return int(p.x), int(-p.y+height)
def from_pygame(p):
return to_pygame(p)
clock = pygame.time.Clock()
running = True
font = pygame.font.Font(None, 16)
# Create the torso box.
box_width = 50
box_height = 100
# leg_length = 100
leg_length = 125
leg_thickness = 2
leg_shape_filter = pymunk.ShapeFilter(group=LEG_GROUP)
# Create torso.
mass = 200
points = [(-box_width/2, -box_height/2), (-box_width/2, box_height/2), (box_width/2, box_height/2), (box_width/2, -box_height/2)]
moment = pymunk.moment_for_poly(mass, points)
body1 = pymunk.Body(mass, moment)
body1.position = (self.display_size[0]/2, self.ground_y+box_height/2+leg_length)
body1.start_position = Vec2d(body1.position)
body1.start_angle = body1.angle
shape1 = pymunk.Poly(body1, points)
shape1.filter = leg_shape_filter
shape1.friction = 0.8
shape1.elasticity = 0.0
self.space.add(body1, shape1)
# Create leg extending from the right to the origin.
mass = 10
points = [
(leg_thickness/2, -leg_length/2),
(-leg_thickness/2, -leg_length/2),
(-leg_thickness/2, leg_length/2),
(leg_thickness/2, leg_length/2)
]
moment = pymunk.moment_for_poly(mass, points)
body2 = pymunk.Body(mass, moment)
body2.position = (self.display_size[0]/2-box_width/2+leg_thickness/2, self.ground_y+leg_length/2)
body2.start_position = Vec2d(body2.position)
body2.start_angle = body2.angle
shape2 = pymunk.Poly(body2, points)
shape2.filter = leg_shape_filter
shape2.friction = 0.8
shape2.elasticity = 0.0
self.space.add(body2, shape2)
# Link bars together at end.
pj = pymunk.PivotJoint(body1, body2, (self.display_size[0]/2-box_width/2, self.ground_y+leg_length))
self.space.add(pj)
# Attach the foot to the ground in a fixed position.
# We raise it above by the thickness of the leg to simulate a ball-foot. Otherwise, the default box foot creates discontinuities.
pj = pymunk.PivotJoint(self.space.static_body, body2, (self.display_size[0]/2-box_width/2, self.ground_y+leg_thickness))
self.space.add(pj)
# Actuate the bars via a motor.
motor_joint = pymunk.SimpleMotor(body1, body2, 0)
motor_joint.max_force = 1e10 # mimicks default infinity
# motor_joint.max_force = 1e9
# motor_joint.max_force = 1e7 # too weak, almost no movement
self.space.add(motor_joint)
# Add hard stops to leg pivot so the leg can't rotate through the torso.
hip_limit_joint = pymunk.RotaryLimitJoint(body1, body2, -pi/4., pi/4.) # -45deg:+45deg
self.space.add(hip_limit_joint)
last_body1_pos = None
last_body1_vel = None
simulate = False
while running:
# print('angles:', body1.angle, body2.angle)
# print('torso force:', body1.force)
print('body1.position: %.02f %.02f' % (body1.position.x, body1.position.y))
current_body1_vel = None
if last_body1_pos:
current_body1_vel = body1.position - last_body1_pos
print('current_body1_vel: %.02f %.02f' % (current_body1_vel.x, current_body1_vel.y))
current_body1_accel = None
if last_body1_vel:
current_body1_accel = current_body1_vel - last_body1_vel
print('current_body1_accel: %.02f %.02f' % (current_body1_accel.x, current_body1_accel.y))
servo_angle = (body1.angle - body2.angle) * 180/pi # 0 degrees means leg is angled straight down
servo_cw_enabled = servo_angle > -45
servo_ccw_enabled = servo_angle < 45
for event in pygame.event.get():
if event.type == QUIT or (event.type == KEYDOWN and event.key in (K_q, K_ESCAPE)):
sys.exit(0)
elif event.type == KEYDOWN and event.key == K_s:
# Start/stop simulation.
simulate = not simulate
last_body1_pos = Vec2d(body1.position)
if current_body1_vel:
last_body1_vel = Vec2d(current_body1_vel)
self.draw()
### Update physics
fps = 50
iterations = 25
dt = 1.0/float(fps)/float(iterations)
if simulate:
for x in range(iterations): # 10 iterations to get a more stable simulation
self.space.step(dt)
pygame.display.flip()
clock.tick(fps)
if __name__ == '__main__':
sim = Simulator()
sim.main()
This renders a box placed on top of a thin leg. The leg is connected to the box by a pivot joint, and to the ground via another pivot joint. However, the leg is attached to the box off-center to the left, so the center-of-gravity is unbalanced. In the real world, this setup would cause the box to toppled to the right. However, when you run this code (and press "s" to start), it shows the box toppling to the left. Why is this?
I've tried adjusting the mass (high mass for the box, low mass for the leg), the center of gravity for the box, and tweaking the attachment points for the joints, but nothing seems to change the outcome. What am I doing wrong?
I want to use this for simulating a real-world phenomena, but until I can get it to reproduce the real-world phenomena, I'm stuck.
It seems to be because the leg shape collides with the bottom ground shape.
The easiest way to make them not collide is to move them apart a little bit. For example make the leg a little shorter so that it doesnt touch the ground.
Another solution is to do as you did in your other question, ignore collisions between the leg and the ground. To do that you can setup a shape filter, but since you probably want to keep the box from colliding with the leg, and at the same time count collisions between the box and ground I think you need to use the categories/masks of the shape filter as documented here: http://www.pymunk.org/en/latest/pymunk.html#pymunk.ShapeFilter
I made a program using Python, with pygame, that loads pictures of materials and then creates blocks and each block is assigned with random material.
Block is a class and in the drawing process, it iterates through the array with stored blocks, but that is very slow. Isn't there a faster method than storing them in array and iterating through?
class block:
def __init__(self, texture, x, y):
self.texture = texture
self.x = x
self.y = y
material = pygame.image
material.grass = pygame.image.load("textures/grass.png")
material.water = pygame.image.load("textures/water.png")
material.sand = pygame.image.load("textures/sand.png")
materials = [material.grass, material.water, material.sand]
white = (255,255,255);(width, height) = (2048, 1008);black = (0, 0, 0);screen = pygame.display.set_mode((width, height))
b_unit = 16
b = []
count = 0
cx = 0
cy = 0
while count < (width * height) / (b_unit * b_unit):
b.append(block(random.choice(materials), b_unit * cx, b_unit * cy))
cx += 1
count += 1
if cx == width / b_unit:
cx = 0
cy += 1
while True:
for block in b:
screen.blit(block.texture, (block.x + viewx, block.y + viewy))
pygame.display.flip()
I've already mentioned in the comments that you should (almost) always convert your images to improve the performance.
It can also help to blit separate images/pygame.Surfaces onto a big background surface and then just blit this background once per frame. I use two nested for loops here to get the coordinates and randomly blit one of two images.
I get around 120 fps if I use separate sprites (5184) here and ~430 fps with this single background image.
Of course I'm just blitting here and in a real game you'd probably have to store the rects of the tiles in a list or use pygame sprites and sprite groups, for example to implement collision detection or other map related logic, so the frame rate would be lower.
import itertools
import pygame as pg
from pygame.math import Vector2
BLUE_IMAGE = pg.Surface((20, 20))
BLUE_IMAGE.fill(pg.Color('lightskyblue2'))
GRAY_IMAGE = pg.Surface((20, 20))
GRAY_IMAGE.fill(pg.Color('slategray4'))
def main():
screen = pg.display.set_mode((1920, 1080))
clock = pg.time.Clock()
all_sprites = pg.sprite.Group()
images = itertools.cycle((BLUE_IMAGE, GRAY_IMAGE))
background = pg.Surface(screen.get_size())
# Use two nested for loops to get the coordinates.
for y in range(screen.get_height()//20):
for x in range(screen.get_width()//20):
# This alternates between the blue and gray image.
image = next(images)
# Blit one image after the other at their respective coords.
background.blit(image, (x*20, y*20))
next(images)
done = False
while not done:
for event in pg.event.get():
if event.type == pg.QUIT:
done = True
# Now you can just blit the background image once
# instead of blitting thousands of separate images.
screen.blit(background, (0, 0))
pg.display.set_caption(str(clock.get_fps()))
pg.display.flip()
clock.tick(1000)
if __name__ == '__main__':
pg.init()
main()
pg.quit()
Side notes: Don't add your images to the pygame.image module (that makes no sense at all).
material = pygame.image
material.grass = pygame.image.load("textures/grass.png")
Writing several statements in the same row separated with semicolons is really ugly and makes code less readable.
white = (255,255,255);(width, height) = (2048, 1008)
I have a dynamic object to which I set different velocity values. However when this dynamic body hits a static body, it overlaps partly with the static shape until this collision is resolved and it moves back.
Is there a way in pymunk to make the dynamic body stop exactly at the borders of the static body, even when velocity is applied in this direction? If there are collision conflicts I would rather have them solved in another way than to make the two shapes overlap.
Applying forces and impulses are not really an option since I want to have a constant velocity.
(The below code needs to be executed twice to work.)
import pymunk
import pyglet
from PIL import Image
from PIL import ImageDraw
# setup of pyglet
window = pyglet.window.Window()
main_batch = pyglet.graphics.Batch()
keys = pyglet.window.key.KeyStateHandler()
window.push_handlers(keys)
# setup of pymunk
space = pymunk.Space()
"""MOVABLE CIRCLE"""
# creating pyglet sprite
circle_img = Image.new('RGBA', (50,50))
draw = ImageDraw.Draw(circle_img)
draw.ellipse((1, 1, 50-1, 50-1), fill=(255,0,0))
circle_img.save('circle.png')
pyglet_circle_img = pyglet.resource.image('circle.png')
pyglet_circle_img.anchor_x = pyglet_circle_img.width/2
pyglet_circle_img.anchor_y = pyglet_circle_img.height/2
circle_sprite = pyglet.sprite.Sprite(pyglet_circle_img, window.width/2, window.height/2, batch=main_batch)
# creating pymunk body and shape
mass = 2
radius = 25
moment = pymunk.moment_for_circle(mass, 0, radius)
circle_body = pymunk.Body(mass, moment)
circle_body.position = circle_sprite.position
circle_shape = pymunk.Circle(circle_body, 25)
circle_shape.elasticity = 0.0
space.add(circle_body, circle_shape)
"""STATIC SQUARE"""
# creating pyglet sprite
square_img = Image.new('RGBA', (70,70))
draw = ImageDraw.Draw(square_img)
draw.rectangle([(0, 0), (70-1, 70-1)], fill=(0,255,0))
square_img.save('square.png')
pyglet_square_img = pyglet.resource.image('square.png')
pyglet_square_img.anchor_x = pyglet_square_img.width/2
pyglet_square_img.anchor_y = pyglet_square_img.height/2
square_sprite = pyglet.sprite.Sprite(pyglet_square_img, 3*window.width/4, window.height/2, batch=main_batch)
# creating pymunk body and shape
square_body = pymunk.Body(body_type=pymunk.Body.KINEMATIC)
square_body.position = square_sprite.position
square_shape = pymunk.Poly(square_body, [(-35,-35),(-35,35),(35,35),(35,-35)])
square_shape.elasticity = 0.0
space.add(square_body, square_shape)
def update(dt):
space.step(dt)
circle_sprite.position = circle_body.position
print(circle_body.position)
key_pressed = False
if keys[pyglet.window.key.LEFT]:
circle_body.velocity = (-100,0)
key_pressed = True
elif keys[pyglet.window.key.RIGHT]:
circle_body.velocity = (100, 0)
key_pressed = True
if keys[pyglet.window.key.UP]:
circle_body.velocity = (0, 100)
key_pressed = True
elif keys[pyglet.window.key.DOWN]:
circle_body.velocity = (0, -100)
key_pressed = True
if not key_pressed:
circle_body.velocity = (0,0)
#window.event
def on_draw():
window.clear()
main_batch.draw()
pyglet.clock.schedule_interval(update, 1/60.)
pyglet.app.run()
In general the answer it that you should expect strange effects when you set the position or velocity manually of a body, since that it tricking the physics engine (just like you would get strange effects in real life if
something teleported around)
Its also the case that a small overlap is to be expected when objects collide, that is how the collision is solved. There are two properties on the space that you can use to control this a bit, collision_slop and collision_bias.
However, you can experiment with some manual fixes which might help. One way would be to move the objects away once a collision happen. You can do this with a collision callback.
Here is a quick example you can put just before the update function in your example to prevent the overlap:
circle_shape.collision_type = 1
h = space.add_wildcard_collision_handler(circle_shape.collision_type)
def f(arbiter, space, data):
ps = arbiter.contact_point_set
arbiter.shapes[0].body.position += ps.normal * ps.points[0].distance
h.post_solve = f
(in your real code you need to add some fail safes and also account for more than one point in case you have more complex shapes)