bob_llm.llm_node

Classes

ROS 2 node that provides an interface to LLMs and VLMs.

Functions

main([args])

Module Contents

class bob_llm.llm_node.LLMNode

Bases: rclpy.node.Node

ROS 2 node that provides an interface to LLMs and VLMs.

This node handles prompts, manages conversation history, and executes tools.

chat_history = []

_prefix_history_len

_is_generating = False

_cancel_requested = False

_prompt_queue

_queue_timer = None

_queue_timer_period = 0.1

sub

pub_response

pub_stream

pub_latest_turn

_initialize_chat_history(): Populate the initial chat history from ROS parameters.

load_llm_client(): Load and configure the LLM client based on ROS parameters.

_load_tools()

Dynamically load tool modules specified in ‘tool_interfaces’.

Returns:: A tuple containing (all_tools, all_functions).

_publish_latest_turn(user_prompt: str, assistant_message: dict)

Process the latest conversational turn for publishing and logging.

Parameters:

user_prompt – The string content of the user’s latest prompt.
assistant_message – The final message dictionary from the assistant.

_get_truncated_history(): Return a copy of chat history with long strings truncated.

_trim_chat_history()

Prevent the chat history from growing indefinitely.

It trims the oldest conversational turns to stay within the limit.

prompt_callback(msg): Process an incoming prompt from the ‘llm_prompt’ topic.

_process_queue_callback(): Timer callback to process queued prompts.

bob_llm.llm_node.main(args=None)