Agent Engine

Bases: ABC

Abstract base class representing the engine that manages agents.

The engine is responsible for creating, managing, and terminating agents, as well as facilitating interactions between them.

create_agent(name, **kwargs) abstractmethod

Instantiate a new agent within the engine.

Parameters:

Name	Type	Description	Default
name	str	The unique name identifier for the agent.	required
**kwargs	Dict[str, Any]	Additional keyword arguments for agent initialization.	{}

Returns:

Name	Type	Description
IAgent	IAgent	The newly created agent instance.

Raises:

Type	Description
ValueError	If an agent with the same name already exists.
KeyError	If start_node_id is not provided in kwargs.

create_iter() abstractmethod

Create an iterator for processing agent steps.

Returns:

Type	Description
Iterable[IAgent]	Iterable[IAgent]: An iterator object over all agents.

delete_agent(name) abstractmethod

Delete an agent by its name.

Parameters:

Name	Type	Description	Default
name	str	The unique name identifier of the agent to retrieve.	required

Returns:

Type	Description
None	None

Raises:

Type	Description
KeyError	If no agent with the specified name exists.

get_agent(name) abstractmethod

Retrieve an agent by its name.

Parameters:

Name	Type	Description	Default
name	str	The unique name identifier of the agent to retrieve.	required

Returns:

Name	Type	Description
IAgent	IAgent	The agent instance corresponding to the provided name.

Raises:

Type	Description
KeyError	If no agent with the specified name exists.

terminate() abstractmethod

Terminate the agent engine and perform necessary cleanup operations.

This method should ensure that all resources are properly released and that agents are gracefully shut down.

Agent

---

Bases: ABC

Abstract base class representing an agent in the system.

current_node_id abstractmethod property

Get the current node ID of the agent.

name abstractmethod property

Get the name identifier of the agent.

orientation abstractmethod property

Get the orientation of the agent.

Returns:

Name	Type	Description
int	Tuple[float, float]	The current orientation of the agent.

prev_node_id abstractmethod property

Get the previous node ID of the agent.

state abstractmethod property

Get the current state of the agent.

Returns:

Type	Description
Dict[str, Any]	Dict[str, Any]: The current state data of the agent.

strategy abstractmethod property

Get the current strategy of the agent.

Returns:

Type	Description
Optional[Callable[[Dict[str, Any]], None]]	Optional[Callable[[Dict[str, Any]], None]]: The current strategy function or None if no strategy is set.

deregister_sensor(name) abstractmethod

Deregister a sensor from the agent.

Parameters:

Name	Type	Description	Default
name	str	The unique name identifier for the sensor to be deregistered.	required

get_state() abstractmethod

Retrieve the current state of the agent.

Returns:

Type	Description
Dict[str, Any]	Dict[str, Any]: The current state data of the agent, structure depends on implementation.

register_sensor(name, sensor) abstractmethod

Register a sensor with the agent.

Sensors can be used by the agent to perceive the environment or gather data.

Parameters:

Name	Type	Description	Default
name	str	The unique name identifier for the sensor.	required
sensor	ISensor	The sensor instance or object to be registered.	required

register_strategy(strategy) abstractmethod

Register a strategy with the agent.

Strategies define the decision-making or action-planning mechanisms for the agent.

Parameters:

Name	Type	Description	Default
strategy	Callable[[Dict[str, Any]], None]	The strategy instance or object to be registered.	required

set_state() abstractmethod

Update the agent's state.

step() abstractmethod

Execute a single operational step of the agent.

This method should contain the logic that defines the agent's behavior during one iteration or time step in the system.