def generate_nn_polyhedral_guard(self, nn, chosen_action, output_flag):
gurobi_model = grb.Model()
gurobi_model.setParam('OutputFlag', output_flag)
gurobi_model.setParam('Threads', 2)
observation = gurobi_model.addMVar(shape=(2, ),
lb=float("-inf"),
ub=float("inf"),
name="observation")
Experiment.generate_nn_guard(gurobi_model,
observation,
nn,
action_ego=chosen_action)
observable_template = Experiment.octagon(2)
# self.env_input_size = 2
observable_result = optimise(observable_template, gurobi_model,
observation)
# self.env_input_size = 6
return observable_template, observable_result
def post_milp(self, x, x_label, nn, output_flag, t, template):
"""milp method"""
post = []
for chosen_action in range(2):
gurobi_model = grb.Model()
gurobi_model.setParam('OutputFlag', output_flag)
input = generate_input_region(gurobi_model, template, x,
self.env_input_size)
# gurobi_model.addConstr(input[0] >= 0, name=f"input_base_constr1")
# gurobi_model.addConstr(input[1] >= 0, name=f"input_base_constr2")
# gurobi_model.addConstr(input[2] >= 20, name=f"input_base_constr3")
observation = gurobi_model.addMVar(shape=(2, ),
lb=float("-inf"),
ub=float("inf"),
name="observation")
gurobi_model.addConstr(
observation[1] <= input[0] - input[1] + self.input_epsilon / 2,
name=f"obs_constr21")
gurobi_model.addConstr(
observation[1] >= input[0] - input[1] - self.input_epsilon / 2,
name=f"obs_constr22")
gurobi_model.addConstr(observation[0] <= self.v_lead - input[2] +
self.input_epsilon / 2,
name=f"obs_constr11")
gurobi_model.addConstr(observation[0] >= self.v_lead - input[2] -
self.input_epsilon / 2,
name=f"obs_constr12")
# gurobi_model.addConstr(input[3] <= self.max_speed, name=f"v_constr_input")
# gurobi_model.addConstr(input[3] >= -self.max_speed, name=f"v_constr_input")
feasible_action = Experiment.generate_nn_guard(
gurobi_model, observation, nn, action_ego=chosen_action)
# feasible_action = Experiment.generate_nn_guard(gurobi_model, input, nn, action_ego=chosen_action)
if feasible_action:
# apply dynamic
x_prime = StoppingCarExperiment.apply_dynamic(
input,
gurobi_model,
action=chosen_action,
env_input_size=self.env_input_size)
gurobi_model.update()
gurobi_model.optimize()
found_successor, x_prime_results = self.h_repr_to_plot(
gurobi_model, template, x_prime)
if found_successor:
# post.append((tuple(x_prime_results),(x, x_label)))
successor_info = Experiment.SuccessorInfo()
successor_info.successor = tuple(x_prime_results)
successor_info.parent = x
successor_info.parent_lbl = x_label
successor_info.t = t + 1
successor_info.action = "policy" # chosen_action
# successor_info.lb = ranges_probs[chosen_action][0]
# successor_info.ub = ranges_probs[chosen_action][1]
post.append(successor_info)
return post
def post_milp(self, x, x_label, nn, output_flag, t,
template) -> List[Experiment.SuccessorInfo]:
"""milp method"""
ranges_probs = self.create_range_bounds_model(template, x,
self.env_input_size, nn)
post = []
for chosen_action in range(2):
gurobi_model = grb.Model()
gurobi_model.setParam('OutputFlag', output_flag)
gurobi_model.setParam('DualReductions', 0)
input = generate_input_region(gurobi_model, template, x,
self.env_input_size)
observation = gurobi_model.addMVar(shape=(2, ),
lb=float("-inf"),
ub=float("inf"),
name="input")
gurobi_model.addConstr(
observation[1] <= input[0] - input[1] + self.input_epsilon / 2,
name=f"obs_constr21")
gurobi_model.addConstr(
observation[1] >= input[0] - input[1] - self.input_epsilon / 2,
name=f"obs_constr22")
gurobi_model.addConstr(observation[0] <= self.v_lead - input[2] +
self.input_epsilon / 2,
name=f"obs_constr11")
gurobi_model.addConstr(observation[0] >= self.v_lead - input[2] -
self.input_epsilon / 2,
name=f"obs_constr12")
feasible_action = Experiment.generate_nn_guard(
gurobi_model, observation, nn, action_ego=chosen_action)
if feasible_action:
x_prime = StoppingCarExperiment.apply_dynamic(
input,
gurobi_model,
action=chosen_action,
env_input_size=self.env_input_size)
gurobi_model.update()
gurobi_model.optimize()
found_successor, x_prime_results = self.h_repr_to_plot(
gurobi_model, template, x_prime)
if found_successor:
successor_info = Experiment.SuccessorInfo()
successor_info.successor = tuple(x_prime_results)
successor_info.parent = x
successor_info.parent_lbl = x_label
successor_info.t = t + 1
successor_info.action = "policy" # chosen_action
successor_info.lb = ranges_probs[chosen_action][0]
successor_info.ub = ranges_probs[chosen_action][1]
post.append(successor_info)
return post
def post_milp(self, x, nn, output_flag, t, template):
"""milp method"""
post = []
for chosen_action in range(2):
gurobi_model = grb.Model()
gurobi_model.setParam('OutputFlag', output_flag)
input = generate_input_region(gurobi_model, template, x,
self.env_input_size)
observation = gurobi_model.addMVar(shape=(2, ),
lb=float("-inf"),
ub=float("inf"),
name="input")
gurobi_model.addConstr(
observation[1] <= input[0] - input[1] + self.input_epsilon / 2,
name=f"obs_constr21")
gurobi_model.addConstr(
observation[1] >= input[0] - input[1] - self.input_epsilon / 2,
name=f"obs_constr22")
gurobi_model.addConstr(observation[0] <= self.v_lead - input[2] +
self.input_epsilon / 2,
name=f"obs_constr11")
gurobi_model.addConstr(observation[0] >= self.v_lead - input[2] -
self.input_epsilon / 2,
name=f"obs_constr12")
feasible_action = Experiment.generate_nn_guard(
gurobi_model, observation, nn, action_ego=chosen_action)
# feasible_action = Experiment.generate_nn_guard(gurobi_model, input, nn, action_ego=chosen_action)
if feasible_action:
# apply dynamic
x_prime = StoppingCarExperiment2.apply_dynamic(
input,
gurobi_model,
action=chosen_action,
env_input_size=self.env_input_size)
gurobi_model.update()
gurobi_model.optimize()
found_successor, x_prime_results = self.h_repr_to_plot(
gurobi_model, template, x_prime)
if found_successor:
post.append(tuple(x_prime_results))
return post