erl_path_planning_ros

This package provides ROS nodes that wrap the erl_path_planning library for use in robotics applications. It includes support for both standard 2D path planning and Linear Temporal Logic (LTL) based path planning for mission specification.

A* 2D	A* 2D with LTL

Features

• A* Path Planning: Optimal path planning using the A* algorithm
• LTL Mission Planning: Support for complex mission specifications using Linear Temporal Logic
• ROS2 Integration: Full integration with ROS2 interfaces
• Flexible Precision: Supports both single and double precision computation
• Configurable Motion Primitives: Customizable grid-based motion with diagonal movements
• Robot Footprint Support: Considers robot geometry for collision checking
• QoS Configuration: Flexible QoS settings for all topics

Nodes

Common Parameters for A* Based Nodes

All A* nodes inherit from a common base class (AstarNode) that provides the following shared parameters:

Core Settings

• double_precision (bool, default: false): Use double precision instead of float
• global_frame (string, default: "map"): Global reference frame
• robot_frame (string, default: "base_link"): Robot reference frame
• start_source (string, default: "topic"): Start source ("topic" or "tf")

Default QoS Settings

• default_qos_reliability (string, default: "best_effort"): Default QoS reliability for topics
• default_qos_durability (string, default: "transient_local"): Default QoS durability for topics

Input Topics

• start_topic (string, default: "start"): Start pose topic
• start_topic_reliability (string, default: uses default_qos_reliability)
• start_topic_durability (string, default: uses default_qos_durability)
• goal_topic (string, default: "goals"): Goal positions topic
• goal_topic_reliability (string, default: uses default_qos_reliability)
• goal_topic_durability (string, default: uses default_qos_durability)
• goal_tolerance_topic (string, default: "goal_tolerances"): Goal tolerances topic
• goal_tolerance_topic_reliability (string, default: uses default_qos_reliability)
• goal_tolerance_topic_durability (string, default: uses default_qos_durability)
• terminal_cost_topic (string, default: "terminal_costs"): Terminal costs topic
• terminal_cost_topic_reliability (string, default: uses default_qos_reliability)
• terminal_cost_topic_durability (string, default: uses default_qos_durability)

Output Topics

• path_topic (string, default: "plan"): Path output topic
• path_topic_reliability (string, default: uses default_qos_reliability)
• path_topic_durability (string, default: uses default_qos_durability)
• cost_topic (string, default: "cost"): Cost output topic
• cost_topic_reliability (string, default: uses default_qos_reliability)
• cost_topic_durability (string, default: uses default_qos_durability)
• goal_idx_topic (string, default: "goal_idx"): Goal index output topic
• goal_idx_topic_reliability (string, default: uses default_qos_reliability)
• goal_idx_topic_durability (string, default: uses default_qos_durability)

Services

• plan_srv_name (string, default: "plan"): Planning service name
• reset_srv_name (string, default: "reset"): Reset service name

A* Algorithm Settings

• eps (double, default: 1.0): Suboptimality factor (1.0 = optimal, >1.0 = faster but suboptimal)
• max_num_iterations (int, default: 100000): Maximum number of search iterations
• log (bool, default: false): Enable verbose logging
• reopen_inconsistent (bool, default: false): Reopen inconsistent nodes in search

---

1. astar_2d_node

Standard 2D A* path planner for grid-based navigation.

Subscriptions

• map (nav_msgs/OccupancyGrid): Occupancy grid map for planning (default topic: "map")
• start (std_msgs/Float64MultiArray): Start position as (x, y) vector (default topic: "start")
• goals (std_msgs/Float64MultiArray): Goal positions as (2, N) matrix (default topic: "goals")
• goal_tolerances (std_msgs/Float64MultiArray): Goal tolerances (default topic: "goal_tolerances")
• terminal_costs (std_msgs/Float64MultiArray): Terminal costs for each goal (default topic: "terminal_costs")

Publications

• plan (nav_msgs/Path): Computed path from start to goal
• cost (std_msgs/Float64): Total cost of the computed path
• goal_idx (std_msgs/Int64): Index of the reached goal

Services

• plan (std_srvs/srv/Trigger): Trigger path planning
• reset (std_srvs/srv/Trigger): Reset the planner state

Parameters

See Common Parameters for A* Based Nodes above.

Additional Node-Specific Parameters:

Map Configuration:

• map_topic (string, default: "map"): Occupancy grid topic
• map_topic_reliability (string, default: "best_effort"): QoS reliability setting
• map_topic_durability (string, default: "transient_local"): QoS durability setting

Motion Primitives:

• max_axis_step (int, default: 1): Maximum step along one axis for motion primitives
• allow_diagonal (bool, default: true): Allow diagonal movements

Robot Configuration:

• robot_metric_contour (double[], default: []): Robot footprint contour as [x1, y1, x2, y2, ...] (minimum 3 points)

Environment Settings:

• obstacle_threshold (int, default: varies): Occupancy threshold for obstacles (0-100)
• add_map_cost (bool, default: false): Add map costs to path cost
• map_cost_factor (double, default: 0.0): Factor to scale map costs

2. astar_2d_ltl_node

A* path planner with Linear Temporal Logic (LTL) mission specification support.

Subscriptions

• occ_map (nav_msgs/OccupancyGrid): Occupancy grid map (default topic: "occ_map")
• label_map (erl_geometry_msgs/GridMapMsg): Label map (UINT32) for atomic propositions (default topic: "label_map")
• fsa_spot (std_msgs/String): Finite State Automaton in Spot HOA format (default topic: "fsa_spot")
• start (std_msgs/Float64MultiArray): Start position (default topic: "start")
• goals (std_msgs/Float64MultiArray): Goal positions (default topic: "goals")
• goal_tolerances (std_msgs/Float64MultiArray): Goal tolerances (default topic: "goal_tolerances")
• terminal_costs (std_msgs/Float64MultiArray): Terminal costs (default topic: "terminal_costs")

Publications

• plan (nav_msgs/Path): Computed path satisfying LTL specification
• cost (std_msgs/Float64): Total cost of the computed path
• goal_idx (std_msgs/Int64): Index of the reached goal
• ap_dict (std_msgs/String): Atomic proposition dictionary in YAML format

Services

• plan (std_srvs/srv/Trigger): Trigger LTL-aware path planning
• reset (std_srvs/srv/Trigger): Reset the planner state

Parameters

See Common Parameters for A* Based Nodes above.

Additional LTL-Specific Topics:

• label_map_topic (string, default: "label_map"): Label map topic
• label_map_topic_reliability (string, default: "best_effort")
• label_map_topic_durability (string, default: "transient_local")
• occ_map_topic (string, default: "occ_map"): Occupancy map topic
• occ_map_topic_reliability (string, default: "best_effort")
• occ_map_topic_durability (string, default: "transient_local")
• fsa_topic (string, default: "fsa_spot"): FSA topic in Spot HOA format
• fsa_topic_reliability (string, default: "reliable")
• fsa_topic_durability (string, default: "transient_local")
• ap_dict_topic (string, default: "ap_dict"): Atomic proposition dictionary output topic

Note: The astar_2d_ltl_node does not use the map_topic parameter. Instead, it uses occ_map_topic and label_map_topic for separate occupancy and label information.

Usage Examples

Basic 2D Planning

# Launch with default parameters
ros2 run erl_path_planning_ros astar_2d_node

# Launch with custom parameters
ros2 run erl_path_planning_ros astar_2d_node --ros-args \
  -p double_precision:=true \
  -p global_frame:=map \
  -p eps:=1.5 \
  -p max_axis_step:=2 \
  -p allow_diagonal:=true \
  -p robot_metric_contour:="[-0.3, -0.3, 0.3, -0.3, 0.3, 0.3, -0.3, 0.3]"

# Publish start and goals
ros2 topic pub /start std_msgs/msg/Float64MultiArray \
  "{layout: {dim: [{size: 2, stride: 1}]}, data: [0.0, 0.0]}"

ros2 topic pub /goals std_msgs/msg/Float64MultiArray \
  "{layout: {dim: [{size: 2, stride: 1}, {size: 1, stride: 1}]}, data: [5.0, 5.0]}"

# Trigger planning
ros2 service call /plan std_srvs/srv/Trigger

LTL-Based Planning

# Launch LTL planner
ros2 run erl_path_planning_ros astar_2d_ltl_node --ros-args \
  -p double_precision:=false \
  -p eps:=1.0 \
  -p max_num_iterations:=200000

# The node will wait for:
# - Occupancy map on /occ_map
# - Label map on /label_map
# - FSA specification on /fsa_spot
# Then publish the atomic proposition dictionary on /ap_dict

# Trigger planning
ros2 service call /plan std_srvs/srv/Trigger

Dependencies

• erl_path_planning: Core path planning algorithms
• erl_env: Environment representations
• erl_geometry_msgs: Provides GridMapMsg
• erl_common: Common utilities
• ROS1 Noetic (TODO)
• ROS2 Humble or later

Building

ROS2

# In your ROS2 workspace
colcon build --packages-select erl_path_planning_ros