#include <tgBulletUnidirComprSpr.h>

Inheritance diagram for tgBulletUnidirComprSpr:

Collaboration diagram for tgBulletUnidirComprSpr:

Public Member Functions
	tgBulletUnidirComprSpr (const std::vector< tgBulletSpringCableAnchor * > &anchors, bool isFreeEndAttached, double coefK, double coefD, double restLength, btVector3 *direction)

virtual	~tgBulletUnidirComprSpr ()

virtual void	step (double dt)

virtual const double	getCurrentAnchorDistanceAlongDirection () const

virtual const double	getCurrentSpringLength () const

virtual const btVector3	getSpringEndpoint () const

virtual const double	getSpringForce () const

virtual const btVector3 *	getDirection () const

virtual const double	getCurrentAnchorDistance () const

virtual const btVector3	getAnchorDirectionUnitVector () const

virtual const double	getCoefK () const

virtual const double	getCoefD () const

virtual const double	getVelocity () const

virtual const double	getDampingForce () const

virtual const double	getRestLength () const

virtual const bool	isFreeEndAttached () const

virtual const std::vector < const tgSpringCableAnchor * >	getAnchors () const

Protected Member Functions
virtual void	calculateAndApplyForce (double dt)

Protected Attributes
btVector3 *	m_direction

std::vector < tgBulletSpringCableAnchor * >	m_anchors

tgBulletSpringCableAnchor *const	anchor1

tgBulletSpringCableAnchor *const	anchor2

double	m_dampingForce

double	m_velocity

bool	m_isFreeEndAttached

const double	m_coefK

const double	m_coefD

double	m_restLength

double	m_prevLength

Detailed Description

This class defines the passive dynamics of a compression spring system, with damping, in the Bullet physics engine. It extends tgBulletCompressionSpring, and changes the way that forces are applied: instead of calculating and applying force along the vector between the two anchor points, it only calculates/applies force along one of the axes: X, Y or Z. So for example, if it's only in Z, one anchor could move around in X or Y and the applied force wouldn't change. Here, we use "Unidir" to mean "Unidirectional."

Definition at line 60 of file tgBulletUnidirComprSpr.h.

Constructor & Destructor Documentation

tgBulletUnidirComprSpr::tgBulletUnidirComprSpr	(	const std::vector< tgBulletSpringCableAnchor * > &	anchors,
		bool	isFreeEndAttached,
		double	coefK,
		double	coefD,
		double	restLength,
		btVector3 *	direction
	)

The only constructor. Takes a list of anchors, a flag controlling attached-ness of the free end of the spring, a coefficient of stiffness, a coefficent of damping, rest length of the spring, and a btVector3 in one of the X, Y, or Z directions.

Parameters

[in]	anchors	- a list of this spring cable's attachements
[in]	isFreeEndAttached	- boolean flag. If no, the spring only provides compression force. If yes, the spring is "attached" to both anchors, and provides a tension force too when the distance between anchors is greater than rest length.
[in]	coefK	- the stiffness of the spring. Must be positive
[in]	dampingCoefficient	- the damping in the spring. Must be non-negative.
[in]	restLength	- the length of the compression spring when unloaded.
[in]	direction	- the direction of the force to be applied, a btVector3. the current version will only support (1,0,0), (0,1,0), or (0,0,1), or the negatives of those vectors. NOTE that if the To anchor is in the negative direction with respect to the From anchor, then direction should be NEGATIVE. NOTE that although direction is a pointer, it is not deleted in this class or any of the other related classes in core. That is because it's stored in the const Config struct of an application: it's only created once, not dynamically, so it's only deleted at the very end of the application (NOT during any individual setups or teardowns.)

The main constructor for this class In comparison to the tgSpringCable vs. tgBulletSpringCable class inheritance, there is no need for the tgCast here, since the same type of anchor is used in both tgBulletCompressionSpring and tgBulletUnidirComprSpr.

Definition at line 43 of file tgBulletUnidirComprSpr.cpp.

tgBulletUnidirComprSpr::~tgBulletUnidirComprSpr ( )

virtual

The virtual destructor. Deletes all of the anchors including anchor1 and anchor2

Definition at line 83 of file tgBulletUnidirComprSpr.cpp.

Member Function Documentation

void tgBulletUnidirComprSpr::calculateAndApplyForce ( double dt )

protectedvirtual

Calculates the current forces that need to be applied to the rigid bodies, and applies them to the bodies of anchor1 and anchor2 Need to re-declare it here so it can be redefined in this child class.

This is the method that does the heavy lifting in this class. Called by step, it calculates what force the spring is experiencing, and applies that force to the rigid bodies it connects to.

Reimplemented from tgBulletCompressionSpring.

Definition at line 229 of file tgBulletUnidirComprSpr.cpp.

Here is the caller graph for this function:

const btVector3 tgBulletCompressionSpring::getAnchorDirectionUnitVector ( ) const

virtualinherited

Return the unit vector in the direction of this spring

Returns the unit vector in the direction of this spring.

Definition at line 181 of file tgBulletCompressionSpring.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

const std::vector< const tgSpringCableAnchor * > tgBulletCompressionSpring::getAnchors ( ) const

virtualinherited

Returns a const vector of const anchors. Currently casts from tgBulletSpringCableAnchors, which makes it impossible to return a reference

Todo:: figure out how to cast and pass by reference

Definition at line 296 of file tgBulletCompressionSpring.cpp.

Here is the caller graph for this function:

virtual const double tgBulletCompressionSpring::getCoefD ( ) const

inlinevirtualinherited

Get the coefficent of damping

Definition at line 124 of file tgBulletCompressionSpring.h.

Here is the caller graph for this function:

virtual const double tgBulletCompressionSpring::getCoefK ( ) const

inlinevirtualinherited

Get the coefficent of stiffness

Definition at line 116 of file tgBulletCompressionSpring.h.

Here is the caller graph for this function:

const double tgBulletCompressionSpring::getCurrentAnchorDistance ( ) const

virtualinherited

Finds the distance between anchor1 and anchor2, and returns the length between them

Returns the distance between the two anchors. This is similar to the getActualLength function in tgBulletSpringCable.

Definition at line 139 of file tgBulletCompressionSpring.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

const double tgBulletUnidirComprSpr::getCurrentAnchorDistanceAlongDirection ( ) const

virtual

Returns the distance between anchors along the vector m_direction. This function just calculates the dot product between m_direction and getCurrentAnchorDistance from the parent class.

Dot getCurrentAnchorDistance with m_direction.

Definition at line 124 of file tgBulletUnidirComprSpr.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

const double tgBulletUnidirComprSpr::getCurrentSpringLength ( ) const

virtual

Returns either restLength or the length of the spring along the single direction, depending on isFreeEndAttached.

Returns the current length of the spring. If isFreeEndAttached, this can be either greater or less than m_restLength. If not, then spring can only exist in compression (less than m_restLength). This only calculates the distance in the stated direction.

Reimplemented from tgBulletCompressionSpring.

Definition at line 142 of file tgBulletUnidirComprSpr.cpp.

virtual const double tgBulletCompressionSpring::getDampingForce ( ) const

inlinevirtualinherited

Get the last value of the damping force

Definition at line 140 of file tgBulletCompressionSpring.h.

virtual const btVector3* tgBulletUnidirComprSpr::getDirection ( ) const

inlinevirtual

Return the direction of applied force for this spring

Definition at line 138 of file tgBulletUnidirComprSpr.h.

virtual const double tgBulletCompressionSpring::getRestLength ( ) const

inlinevirtualinherited

Get the rest length of the spring (we're using these accessor functions per C++ style guidelines.)

Definition at line 149 of file tgBulletCompressionSpring.h.

const btVector3 tgBulletUnidirComprSpr::getSpringEndpoint ( ) const

virtual

Return the location of the free end of the spring. This is used, in particular, for rendering. See tgBulletRenderer.

Returns the location of the endpoint of the spring in space. The renderer uses this to draw lines more easily.

Reimplemented from tgBulletCompressionSpring.

Definition at line 174 of file tgBulletUnidirComprSpr.cpp.

const double tgBulletUnidirComprSpr::getSpringForce ( ) const

virtual

Returns the force currently in the spring, either compression only (e.g., a positive force only if isFreeEndAttached is false), or potentially either + or - force (if isFreeEndAttached == true).

Returns the current force in the spring. If ~isFreeEndAttached, this is zero if the distance between anchors is larger than restLength. Note that this does NOT include the force due to the damper, since that force is only temporary and isn't reallyt "the force in the spring" per se.

Reimplemented from tgBulletCompressionSpring.

Definition at line 194 of file tgBulletUnidirComprSpr.cpp.

Here is the caller graph for this function:

virtual const double tgBulletCompressionSpring::getVelocity ( ) const

inlinevirtualinherited

Get the last change in length / time

Definition at line 132 of file tgBulletCompressionSpring.h.

virtual const bool tgBulletCompressionSpring::isFreeEndAttached ( ) const

inlinevirtualinherited

Return the boolean: is the free end attached?

Definition at line 157 of file tgBulletCompressionSpring.h.

void tgBulletUnidirComprSpr::step ( double dt )

virtual

Updates this object. Calls calculateAndApplyForce(dt)

Parameters

[in] dt,must be positive

Reimplemented from tgBulletCompressionSpring.

Definition at line 91 of file tgBulletUnidirComprSpr.cpp.

Here is the call graph for this function:

Member Data Documentation

tgBulletSpringCableAnchor* const tgBulletCompressionSpring::anchor1

protectedinherited

The first attachement point for this spring cable. Storing it seperately makes a number of functions easier

Definition at line 186 of file tgBulletCompressionSpring.h.

tgBulletSpringCableAnchor* const tgBulletCompressionSpring::anchor2

protectedinherited

The other permanent attachment for this spring cable.

Definition at line 191 of file tgBulletCompressionSpring.h.

std::vector<tgBulletSpringCableAnchor*> tgBulletCompressionSpring::m_anchors

protectedinherited

The list of contact points. tgBulletSpringCable typically has two whereas tgBulletContactSpringCable will have more. Needs to be stored here for consistent rendering. Vector has the convienence of tgCast functions, and we used to need a random iterator to sort

Definition at line 180 of file tgBulletCompressionSpring.h.

const double tgBulletCompressionSpring::m_coefD

protectedinherited

The damping coefficient. Units of mass / sec. Must be non-negative

Definition at line 222 of file tgBulletCompressionSpring.h.

const double tgBulletCompressionSpring::m_coefK

protectedinherited

The stiffness coefficient Units of mass / sec ^2 Must be positive

Definition at line 215 of file tgBulletCompressionSpring.h.

double tgBulletCompressionSpring::m_dampingForce

protectedinherited

The force in the spring due to damping, at the last update step. Stored so we can get it without passing a dt