Only take targets driving roughly in the same direction as IDM target

This commit is contained in:
Johannes Fischer
2022-02-28 11:37:08 +01:00
parent 51eb810f5c
commit 6068c87402

View File

@@ -14,9 +14,9 @@ class PControllerPolicy(BaseAlgorithm):
self._env = env self._env = env
self.target_v = 8.94 # m/s self.target_v = 8.94 # m/s
self.attn_weight = 20 self.attn_weight = 20
# BaseAlgorithm abstract methods # BaseAlgorithm abstract methods
def _setup_model(self): def _setup_model(self):
return None return None
def learn(self, *args, **kwargs): def learn(self, *args, **kwargs):
return self return self
@@ -26,7 +26,7 @@ class PControllerPolicy(BaseAlgorithm):
Generate action, state from observation Generate action, state from observation
(But actually generate next action from underlying environment state) (But actually generate next action from underlying environment state)
Args: Args:
observation (np.ndarray): instantaneous observation from environment observation (np.ndarray): instantaneous observation from environment
@@ -36,16 +36,16 @@ class PControllerPolicy(BaseAlgorithm):
""" """
agent = self._env._agent agent = self._env._agent
ego_state = self._env._env.projected_state[agent].numpy() # (5,) tensor ego_state = self._env._env.projected_state[agent].numpy() # (5,) tensor
# relative_state = np.delete(self._env._env.relative_state[agent].numpy(), agent, axis=0) #(nv-1, 6) tensor # relative_state = np.delete(self._env._env.relative_state[agent].numpy(), agent, axis=0) #(nv-1, 6) tensor
# calculate front and left distances from ego # calculate front and left distances from ego
# calculate relative speed in direction of position difference vector # calculate relative speed in direction of position difference vector
# calculate angle alpha and distance d of vehicle i from ego heading # calculate angle alpha and distance d of vehicle i from ego heading
# attn[i] ~= exp( -(alpha[i])^2 - .01 * d[i] - .1 * vrel[i] # attn[i] ~= exp( -(alpha[i])^2 - .01 * d[i] - .1 * vrel[i]
@@ -60,14 +60,14 @@ class IDMRulePolicy(BaseAlgorithm):
The front car is chosen as the closer of: The front car is chosen as the closer of:
- closest car within a 45 degree half angle cone of the ego's heading - closest car within a 45 degree half angle cone of the ego's heading
- ''' after propagating the environment forward by `t_future' seconds with - ''' after propagating the environment forward by `t_future' seconds with
current headings and velocities current headings and velocities
""" """
def __init__(self, env: Intersimple, def __init__(self, env: Intersimple,
target_speed:float= 8.94, target_speed:float= 8.94,
t_future:List[float]=[0., 1., 2., 3.], t_future:List[float]=[0., 1., 2., 3.],
half_angle:float=60.): half_angle:float=60.):
""" """
Initialize policy with pointer to environment it will run on and target speed Initialize policy with pointer to environment it will run on and target speed
@@ -82,6 +82,7 @@ class IDMRulePolicy(BaseAlgorithm):
self._env = env self._env = env
self.t_future = t_future self.t_future = t_future
self.half_angle = half_angle self.half_angle = half_angle
self.max_heading_diff = 120
# Default IDM parameters # Default IDM parameters
assert target_speed>0, 'negative target speed' assert target_speed>0, 'negative target speed'
@@ -95,18 +96,18 @@ class IDMRulePolicy(BaseAlgorithm):
np.seterr(invalid='ignore') np.seterr(invalid='ignore')
# BaseAlgorithm abstract methods # BaseAlgorithm abstract methods
def _setup_model(self): def _setup_model(self):
return None return None
def learn(self, *args, **kwargs): def learn(self, *args, **kwargs):
return self return self
def predict(self, observation:np.ndarray, def predict(self, observation:np.ndarray,
*args, **kwargs) -> Tuple[np.ndarray, None]: *args, **kwargs) -> Tuple[np.ndarray, None]:
""" """
Predict action, state from observation Predict action, state from observation
(But actually generate next action from underlying environment state) (But actually generate next action from underlying environment state)
Args: Args:
observation (np.ndarray): instantaneous observation from environment observation (np.ndarray): instantaneous observation from environment
@@ -138,11 +139,11 @@ class IDMRulePolicy(BaseAlgorithm):
if t > 0: if t > 0:
xy2 = xy + t * v * np.vstack((np.cos(psi[:,0]), np.sin(psi[:,0]))).T xy2 = xy + t * v * np.vstack((np.cos(psi[:,0]), np.sin(psi[:,0]))).T
d2, r2, i2 = self.get_ego_dr(agent, xy2, v, psi) d2, r2, i2 = self.get_ego_dr(agent, xy2, v, psi)
# choose closer vehicle (now vs imagined) # choose closer vehicle (now vs imagined)
if d2 < d: if d2 < d:
d, r, i = d2, r2, i2 d, r, i = d2, r2, i2
# Update environment interaction graph with i # Update environment interaction graph with i
if i: if i:
self._env._env._graph._neighbor_dict={agent:[i]} self._env._env._graph._neighbor_dict={agent:[i]}
@@ -155,19 +156,19 @@ class IDMRulePolicy(BaseAlgorithm):
assert (d_des>= self.d_min) assert (d_des>= self.d_min)
action = self.a_max*(1 - (s/self.s_max)**4 - (d_des/d)**2) action = self.a_max*(1 - (s/self.s_max)**4 - (d_des/d)**2)
# normalize action to range if env is a NormalizedActionSpace # normalize action to range if env is a NormalizedActionSpace
if isinstance(self._env, NormalizedActionSpace): if isinstance(self._env, NormalizedActionSpace):
action = self._env._normalize(action) action = self._env._normalize(action)
assert action.shape==(1,) assert action.shape==(1,)
return action return action
def get_ego_dr(self, agent:int, xy: np.ndarray, def get_ego_dr(self, agent:int, xy: np.ndarray,
v: np.ndarray, psi: np.ndarray) -> Tuple[float, float, Optional[int]]: v: np.ndarray, psi: np.ndarray) -> Tuple[float, float, Optional[int]]:
""" """
Return distance and relative speed of closest car within half angle from heading Return distance and relative speed of closest car within half angle from heading
Args: Args:
agent (int): agent index agent (int): agent index
xy (np.ndarray): (nv, 2) x and y positions xy (np.ndarray): (nv, 2) x and y positions
@@ -177,7 +178,7 @@ class IDMRulePolicy(BaseAlgorithm):
Returns: Returns:
d (float): distance to closest vehicle in cone d (float): distance to closest vehicle in cone
r (float): relative speed between the two vehicles r (float): relative speed between the two vehicles
i (Optional[int]): index of closest vehicle, or None i (Optional[int]): index of closest vehicle, or None
""" """
nv, nxy = xy.shape nv, nxy = xy.shape
nv2, nvel = v.shape nv2, nvel = v.shape
@@ -192,8 +193,12 @@ class IDMRulePolicy(BaseAlgorithm):
dl = (dxys*np.hstack((-np.sin(psi), np.cos(psi)))).sum(-1) # (nv, ) dl = (dxys*np.hstack((-np.sin(psi), np.cos(psi)))).sum(-1) # (nv, )
alpha = to_circle(np.arctan2(dl, df)) alpha = to_circle(np.arctan2(dl, df))
val_idx = np.arange(nv)[(np.abs(alpha) < self.half_angle*np.pi/180) & (np.arange(nv) != agent)] heading_diff = to_circle(psi - psi[agent]).flatten()
val_idx = np.arange(nv)[
(np.abs(alpha) < self.half_angle*np.pi/180) & (np.arange(nv) != agent) & (np.abs(heading_diff) < self.max_heading_diff*np.pi/180)
]
if len(val_idx)==0: if len(val_idx)==0:
i = None i = None
d = float('inf') d = float('inf')
@@ -202,10 +207,10 @@ class IDMRulePolicy(BaseAlgorithm):
idx = np.argmin(ds[val_idx]) # closest car which meets requirements idx = np.argmin(ds[val_idx]) # closest car which meets requirements
i = int(val_idx[idx]) i = int(val_idx[idx])
d = ds[i] d = ds[i]
r = v[i,0]-v[agent,0] r = v[i,0]-v[agent,0]
return d, r, i return d, r, i
def to_circle(x: np.ndarray) -> np.ndarray: def to_circle(x: np.ndarray) -> np.ndarray:
""" """
Casts x (in rad) to [-pi, pi) Casts x (in rad) to [-pi, pi)