def two_agents_without_interaction_common_playground_123(params):
strt_pnt_A = '16'
strt_pnt_B = '23'
end_pnt_A = '23'
end_pnt_B = '16'
P = md.get_meshgrid_points(params)
x, y, z = get_xyz(P)
# Topology
T, S = md.get_simple_topology_for_regular_grid(params, P)
# rewards
R_A = {end_pnt_A: 100}
R_B = {end_pnt_B: 100}
mdp_challenge_A = {'S': S, 'R': R_A, 'T': T, 'P': P}
mdp_challenge_B = {'S': S, 'R': R_B, 'T': T, 'P': P}
dict_mdp_A = md.start_mdp(params, mdp_challenge_A)
dict_mdp_B = md.start_mdp(params, mdp_challenge_B)
optimal_traj_A = md.get_deterministic_trajectory(strt_pnt_A,
dict_mdp_A,
steps=20)
optimal_traj_B = md.get_deterministic_trajectory(strt_pnt_B,
dict_mdp_B,
steps=20)
interpolated_points_A, points_A = md.get_result_trajectories_mdp(
params, optimal_traj_A, P)
interpolated_points_B, points_B = md.get_result_trajectories_mdp(
params, optimal_traj_B, P)
path_A = interpolated_points_A['quadratic']
path_B = interpolated_points_B['quadratic']
return path_A, path_B
def only_mdp(params):
strt_pnt = '16'
P = md.get_meshgrid_points(params)
x, y, z = get_xyz(P)
# Topology
T, S = md.get_simple_topology_for_regular_grid(params, P)
# rewards
R = {'23': 100}
mdp_challenge = {'S': S, 'R': R, 'T': T, 'P': P}
dict_mdp = md.start_mdp(params, mdp_challenge)
reach_set = md.reach_n_steps(strt_pnt,
mdp_challenge,
dict_mdp,
params,
steps=10)
optimal_traj = md.get_deterministic_trajectory(strt_pnt,
dict_mdp,
steps=20)
print_info(dict_mdp, optimal_traj)
plot_topology(x, y, dict_mdp)
plot_arrows_path(x, y, optimal_traj)
scatter_value_function(x, y, np.array(dict_mdp['U']), dict_mdp, R)
plt_show()
import manifold_plotter as mp
import md_pro as md
import matplotlib.pyplot as plt
mygrid = {
"x_grid": 5,
"y_grid": 4,
"x_min": -10,
"x_max": 10,
"y_min": -7.5,
"y_max": 7.5
}
# points
P = md.get_meshgrid_points(**mygrid)
# plot the nodes
mp.plot_patches(P)
plt.show()