smarts.core.controllers.lane_following_controller module
- class smarts.core.controllers.lane_following_controller.LaneFollowingController[source]
A controller that attempts to follow road waypoints along the vehicle’s plan.
Requires the controlled vehicle to have an Ackermann chassis.
- static calculate_lateral_gains(sim, state, vehicle, desired_poles, target_speed)[source]
Update the state lateral gains
- static find_current_lane(wp_paths, vehicle_position)[source]
Find the lane of the vehicle given a set of waypoint paths.
- heading_error = -15
- lateral_error = -35
- classmethod perform_lane_following(sim, agent_id: str, vehicle, controller_state: LaneFollowingControllerState, sensor_state, target_speed: float = 12.5, lane_change: int = 0)[source]
Control a vehicle
- Parameters:
sim – The simulator instance.
agent_id – The id of the agent controlling the vehicle. XXX: unsure why this is needed
vehicle – The vehicle that is to be controlled.
controller_state – The previous controller state from this controller.
sensor_state – The current sensor state from the vehicle’s sensors.
target_speed – The baseline target speed (controller may give more or less regardless.)
lane_change – Lane index offset from vehicle’s current lane.
- static place_poles(A: ndarray, B: ndarray, poles: ndarray) ndarray[source]
Given a linear system described by ẋ = Ax + Bu, compute the gain matrix K such that the closed loop eigenvalues of A - BK are in the desired pole locations.
References
- side_slip_angle = -3
- yaw_rate = -2