cambrian.agents.point

Point agents.

Classes

MjCambrianAgentPoint	This is a hardcoded class which implements the agent as actuated by a forward
MjCambrianAgentPointSeeker	This is an agent which is non-trainable and defines a custom policy which

Module Contents

class MjCambrianAgentPoint(config, name, *, kp=0.75)

Bases: cambrian.agents.agent.MjCambrianAgent2D

This is a hardcoded class which implements the agent as actuated by a forward velocity and a rotational position. In mujoco, to the best of my knowledge, all translational joints are actuated in reference to the _global_ frame rather than the local frame. This means a velocity actuator applied along the x-axis will move the agent along the global x-axis rather than the local x-axis. Therefore, the agent will have 3 actuators: two for x and y global velocities and one for rotational position. From the perspective the calling class (i.e. MjCambrianEnv), this agent has two actuators: a forward velocity and a rotational position. We will calculate the global velocities and rotational position from these two “actuators”.

Todo

Will create an issue on mujoco and see if it’s possible to implement this in xml. The issue right now is that mujoco doesn’t support relative positions for hinge joints, so we have to implement the heading joint as a velocity actuator which is not ideal.

apply_action(action)

Calls the appropriate apply action method based on the heading joint type.

property action_space: gymnasium.spaces.Space

Overrides the base implementation to only have two elements.

class MjCambrianAgentPointSeeker(config, name, *, target, speed=-0.75, distance_threshold=2.0, use_optimal_trajectory=False)

Bases: MjCambrianAgentPoint

This is an agent which is non-trainable and defines a custom policy which acts as a ‘homing’ agent in the maze environment. This agent will attempt to reach a target (which is either randomly placed or another specific agent) by taking actions that best map to the optimal trajectory which is calculated from the maze using bfs (or just by choosing the action which minimizes the distance to the target).

Keyword Arguments:

• target (Optional[str]) – The name of the target agent to home in on. If None, a random free space in the maze will be chosen as the target.

• speed (float) – The speed at which the agent moves. Defaults to -0.75.

• distance_threshold (float) – The distance threshold at which the agent will consider itself to have reached the target. Defaults to 2.0.

• use_optimal_trajectory (bool) – Whether to use the optimal trajectory to the target. Defaults to False.

reset(*args)

Resets the optimal_trajectory.

get_action_privileged(env)

This is a deviation from the standard gym API. This method is similar to step, but it has “privileged” access to information such as the environment. This method can be overridden by agents which are not trainable and need to implement custom step logic.

Parameters:

env (MjCambrianEnv) – The environment that the agent is in. This can be used to get information about the environment.

Returns:

List[float] – The action to take.