class KBRL_RRT(Planner):
def __init__(self,
robot,
environment,
sampler=None,
dist_to_goal=1.0,
max_iterations=5000,
bias=1.0,
psi=None,
screenshot_rate=1,
save_screenshots=False):
heuristic = lambda s, t: robot.get_dist(s, t)
env = environment
if sampler is None:
sampler = UniformSampler(environment.config_range)
if psi is None:
psi = RBF_Kernel(np.ones(len(environment.config_range)) * 1.0)
self.dist_to_goal = dist_to_goal
self.rl_planner = RL_planner(env, heuristic, sampler, bias, psi)
self.robot = robot
self.max_iterations = max_iterations
self.screenshot_rate = screenshot_rate
self.save_screenshots = save_screenshots
def __call__(self, start, goal):
start = np.array(start)
goal = np.array(goal)
goal_test = lambda x: self.robot.get_dist(x, goal
) < self.dist_to_goal
path, data = self.rl_planner.plan(
start,
goal,
goal_test,
screenshot_rate=self.screenshot_rate,
save_screenshot=self.save_screenshots,
max_iterations=self.max_iterations)
data['robot'] = self.robot
return path, data
class KBRL_RRT(Planner):
def __init__(self,
robot,
environment,
sampler = None,
dist_to_goal = 1.0,
max_iterations=5000,
bias = 1.0,
psi = None,
screenshot_rate = 1,
save_screenshots = False):
heuristic = lambda s, t: robot.get_dist(s, t)
env = environment
if sampler is None:
sampler = UniformSampler(environment.config_range)
if psi is None:
psi = RBF_Kernel(np.ones(len(environment.config_range))*1.0)
self.dist_to_goal = dist_to_goal
self.rl_planner = RL_planner(env, heuristic, sampler, bias, psi)
self.robot = robot
self.max_iterations = max_iterations
self.screenshot_rate = screenshot_rate
self.save_screenshots = save_screenshots
def __call__(self, start, goal):
start = np.array(start)
goal = np.array(goal)
goal_test = lambda x: self.robot.get_dist(x, goal) < self.dist_to_goal
path, data = self.rl_planner.plan(start,
goal,
goal_test,
screenshot_rate = self.screenshot_rate,
save_screenshot = self.save_screenshots,
max_iterations = self.max_iterations)
data['robot'] = self.robot
return path, data
def __init__(self,
robot,
environment,
sampler = None,
dist_to_goal = 1.0,
max_iterations=5000,
bias = 1.0,
psi = None,
screenshot_rate = 1,
save_screenshots = False):
heuristic = lambda s, t: robot.get_dist(s, t)
env = environment
if sampler is None:
sampler = UniformSampler(environment.config_range)
if psi is None:
psi = RBF_Kernel(np.ones(len(environment.config_range))*1.0)
self.dist_to_goal = dist_to_goal
self.rl_planner = RL_planner(env, heuristic, sampler, bias, psi)
self.robot = robot
self.max_iterations = max_iterations
self.screenshot_rate = screenshot_rate
self.save_screenshots = save_screenshots
def __init__(self,
robot,
environment,
sampler=None,
dist_to_goal=1.0,
max_iterations=5000,
bias=1.0,
psi=None,
screenshot_rate=1,
save_screenshots=False):
heuristic = lambda s, t: robot.get_dist(s, t)
env = environment
if sampler is None:
sampler = UniformSampler(environment.config_range)
if psi is None:
psi = RBF_Kernel(np.ones(len(environment.config_range)) * 1.0)
self.dist_to_goal = dist_to_goal
self.rl_planner = RL_planner(env, heuristic, sampler, bias, psi)
self.robot = robot
self.max_iterations = max_iterations
self.screenshot_rate = screenshot_rate
self.save_screenshots = save_screenshots