#include <obstacle_distance_gradient.h>

Inheritance diagram for stomp_moveit::cost_functions::ObstacleDistanceGradient:

Public Member Functions
virtual bool	computeCosts (const Eigen::MatrixXd &parameters, std::size_t start_timestep, std::size_t num_timesteps, int iteration_number, int rollout_number, Eigen::VectorXd &costs, bool &validity) override
	computes the state costs by calculating the minimum distance between the robot and an obstacle.
virtual bool	configure (const XmlRpc::XmlRpcValue &config) override
	Sets internal members of the plugin from the configuration data.
virtual void	done (bool success, int total_iterations, double final_cost, const Eigen::MatrixXd &parameters) override
	Called by the Stomp Task at the end of the optimization process.
virtual std::string	getGroupName () const override
virtual std::string	getName () const override
virtual bool	initialize (moveit::core::RobotModelConstPtr robot_model_ptr, const std::string &group_name, XmlRpc::XmlRpcValue &config) override
	Initializes and configures the Cost Function. Calls the configure method and passes the 'config' value.
	ObstacleDistanceGradient ()
virtual bool	setMotionPlanRequest (const planning_scene::PlanningSceneConstPtr &planning_scene, const moveit_msgs::MotionPlanRequest &req, const stomp_core::StompConfiguration &config, moveit_msgs::MoveItErrorCodes &error_code) override
	Stores the planning details which will be used during the costs calculations.
virtual	~ObstacleDistanceGradient ()
Protected Member Functions
bool	checkIntermediateCollisions (const Eigen::VectorXd &start, const Eigen::VectorXd &end, double longest_valid_joint_move)
	Checks for collision between consecutive points by dividing the joint move into sub-moves where the maximum joint motion can not exceed the longest_valid_joint_move value.
Protected Attributes
collision_detection::CollisionRequest	collision_request_
collision_detection::CollisionResult	collision_result_
std::string	group_name_
std::array < moveit::core::RobotStatePtr, 3 >	intermediate_coll_states_
	Used in checking collisions between to consecutive poses.
double	longest_valid_joint_move_
	how far can a joint move in between consecutive trajectory points
double	max_distance_
	maximum distance from at which the trajectory will be penalized
std::string	name_
moveit_msgs::MotionPlanRequest	plan_request_
planning_scene::PlanningSceneConstPtr	planning_scene_
moveit::core::RobotModelConstPtr	robot_model_ptr_
moveit::core::RobotStatePtr	robot_state_

Detailed Description

Definition at line 44 of file obstacle_distance_gradient.h.

Constructor & Destructor Documentation

stomp_moveit::cost_functions::ObstacleDistanceGradient::ObstacleDistanceGradient ( )

Definition at line 40 of file obstacle_distance_gradient.cpp.

stomp_moveit::cost_functions::ObstacleDistanceGradient::~ObstacleDistanceGradient ( ) [virtual]

Definition at line 47 of file obstacle_distance_gradient.cpp.

Member Function Documentation

bool stomp_moveit::cost_functions::ObstacleDistanceGradient::checkIntermediateCollisions	(	const Eigen::VectorXd &	start,
		const Eigen::VectorXd &	end,
		double	longest_valid_joint_move
	)		`[protected]`

Checks for collision between consecutive points by dividing the joint move into sub-moves where the maximum joint motion can not exceed the longest_valid_joint_move value.

Parameters:

start	The start joint pose
end	The end joint pose
longest_valid_joint_move	The maximum distance that the joints are allowed to move before checking for collisions.

Returns:: True if the interval is collision free, false otherwise.

Definition at line 209 of file obstacle_distance_gradient.cpp.

bool stomp_moveit::cost_functions::ObstacleDistanceGradient::computeCosts	(	const Eigen::MatrixXd &	parameters,
		std::size_t	start_timestep,
		std::size_t	num_timesteps,
		int	iteration_number,
		int	rollout_number,
		Eigen::VectorXd &	costs,
		bool &	validity
	)		`[override, virtual]`

computes the state costs by calculating the minimum distance between the robot and an obstacle.

Parameters:

parameters	The parameter values to evaluate for state costs [num_dimensions x num_parameters]
start_timestep	start index into the 'parameters' array, usually 0.
num_timesteps	number of elements to use from 'parameters' starting from 'start_timestep' *
iteration_number	The current iteration count in the optimization loop
rollout_number	index of the noisy trajectory whose cost is being evaluated. *
costs	vector containing the state costs per timestep.
validity	whether or not the trajectory is valid

Returns:: false if there was an irrecoverable failure, true otherwise.

Implements stomp_moveit::cost_functions::StompCostFunction.

Definition at line 131 of file obstacle_distance_gradient.cpp.

bool stomp_moveit::cost_functions::ObstacleDistanceGradient::configure ( const XmlRpc::XmlRpcValue & config ) [override, virtual]

Sets internal members of the plugin from the configuration data.

Parameters:

config The configuration data . Usually loaded from the ros parameter server

Returns:: true if succeeded, false otherwise.

Implements stomp_moveit::cost_functions::StompCostFunction.

Definition at line 67 of file obstacle_distance_gradient.cpp.

void stomp_moveit::cost_functions::ObstacleDistanceGradient::done	(	bool	success,
		int	total_iterations,
		double	final_cost,
		const Eigen::MatrixXd &	parameters
	)		`[override, virtual]`

Called by the Stomp Task at the end of the optimization process.

Parameters:

success	Whether the optimization succeeded
total_iterations	Number of iterations used
final_cost	The cost value after optimizing.
parameters	The parameters generated at the end of current iteration [num_dimensions x num_timesteps]

Reimplemented from stomp_moveit::cost_functions::StompCostFunction.

Definition at line 255 of file obstacle_distance_gradient.cpp.

virtual std::string stomp_moveit::cost_functions::ObstacleDistanceGradient::getGroupName ( ) const [inline, override, virtual]

Reimplemented from stomp_moveit::cost_functions::StompCostFunction.

Definition at line 95 of file obstacle_distance_gradient.h.

virtual std::string stomp_moveit::cost_functions::ObstacleDistanceGradient::getName ( ) const [inline, override, virtual]

Reimplemented from stomp_moveit::cost_functions::StompCostFunction.

Definition at line 100 of file obstacle_distance_gradient.h.

bool stomp_moveit::cost_functions::ObstacleDistanceGradient::initialize	(	moveit::core::RobotModelConstPtr	robot_model_ptr,
		const std::string &	group_name,
		XmlRpc::XmlRpcValue &	config
	)		`[override, virtual]`

Initializes and configures the Cost Function. Calls the configure method and passes the 'config' value.

Parameters:

robot_model_ptr	A pointer to the robot model.
group_name	The designated planning group.
config	The configuration data. Usually loaded from the ros parameter server

Returns:: true if succeeded, false otherwise.

Implements stomp_moveit::cost_functions::StompCostFunction.

Definition at line 52 of file obstacle_distance_gradient.cpp.

bool stomp_moveit::cost_functions::ObstacleDistanceGradient::setMotionPlanRequest	(	const planning_scene::PlanningSceneConstPtr &	planning_scene,
		const moveit_msgs::MotionPlanRequest &	req,
		const stomp_core::StompConfiguration &	config,
		moveit_msgs::MoveItErrorCodes &	error_code
	)		`[override, virtual]`