trapezoidal velocity trajectory

Collaboration diagram for trapezoidal velocity trajectory:

Data Structures
struct	a_trajtrap
	instance structure for trapezoidal velocity trajectory More...

Typedefs
typedef struct a_trajtrap	a_trajtrap

Functions
double	a_trajtrap_gen (a_trajtrap *ctx, double vm, double ac, double de, double p0, double p1, double v0, double v1)
	generate for trapezoidal velocity trajectory
double	a_trajtrap_pos (a_trajtrap const *ctx, double x)
	calculate position for trapezoidal velocity trajectory
double	a_trajtrap_vel (a_trajtrap const *ctx, double x)
	calculate velocity for trapezoidal velocity trajectory
double	a_trajtrap_acc (a_trajtrap const *ctx, double x)
	calculate acceleration for trapezoidal velocity trajectory

Detailed Description

Function Documentation

a_trajtrap_acc()

double a_trajtrap_acc	(	a_trajtrap const *	ctx,
		double	x )

calculate acceleration for trapezoidal velocity trajectory

\[ \ddot{p}(t)=\begin{cases}a,&t\in[0,t_a)\\0,&t\in[t_a,t_d)\\d,&t\in[t_d,T]\end{cases} \]

Parameters

[in]	ctx	points to an instance of trapezoidal velocity trajectory
[in]	x	difference between current time and initial time

Returns

acceleration output

a_trajtrap_gen()

double a_trajtrap_gen	(	a_trajtrap *	ctx,
		double	vm,
		double	ac,
		double	de,
		double	p0,
		double	p1,
		double	v0,
		double	v1 )

generate for trapezoidal velocity trajectory

Assuming that there is no constant velocity phase, but only acceleration and deceleration phases, the maximum velocity in the motion is \(v\), then we have

\[\frac{v^2-v_0^2}{2a}+\frac{v_1^2-v^2}{2d}=p_1-p_0\]

Solving for the maximum velocity is

\[|v|=\sqrt{\frac{av_1^2-dv_0^2-2ad(p_1-p_0)}{a-d}}\]

1. If \(|v|>|v_m|\), then there is a constant velocity phase.

   \[p_a=p_0+v_0T_a+\frac{1}{2}aT_a^2\]

   \[a_d=p_1-v_cT_d-\frac{1}{2}dT_d^2\]

   \begin{cases}T_a=\cfrac{v_m-v_0}{a}\\T_c=\cfrac{p_d-p_a}{v_c}\\T_d=\cfrac{v_1-v_m}{d}\end{cases}

2. Otherwise there is no constant velocity phase.
   a. If \(|v_0|<|v|\le|v_1|\), there is only an acceleration phase.

   \[|v_1|=\sqrt{v_0^2+2a(p_1-p_0)}\]

   \begin{cases}T_a=\cfrac{v_1-v_0}{a}\\T_c=0\\T_d=0\end{cases}

   b. If \(|v_0|\ge|v|>|v_1|\), there is only a deceleration phase.

   \[|v_1|=\sqrt{v_0^2+2d(p_1-p_0)}\]

   \begin{cases}T_a=0\\T_c=0\\T_d=\cfrac{v_1-v_0}{d}\end{cases}

   c. If \(|v|>|v_0|\), \(|v|>|v_1|\), then there are acceleration and deceleration phases.

   \begin{cases}T_a=\cfrac{v-v_0}{a}\\T_c=0\\T_d=\cfrac{v_1-v}{d}\end{cases}

3. Finally, the position and velocity are calculated using the formula.

   Parameters

   [in,out]	ctx	points to an instance of trapezoidal velocity trajectory
   [in]	vm	defines the maximum velocity during system operation
   [in]	ac	defines the acceleration before constant velocity
   [in]	de	defines the acceleration after constant velocity
   [in]	p0	defines the initial position
   [in]	p1	defines the final position
   [in]	v0	defines the initial velocity
   [in]	v1	defines the final velocity

   Returns

   total duration

a_trajtrap_pos()

double a_trajtrap_pos	(	a_trajtrap const *	ctx,
		double	x )

calculate position for trapezoidal velocity trajectory

\[ p(t)=\begin{cases}v_0+\frac{1}{2}at^2,&t\in[0,t_a)\\p_a+v_c(t-t_a),&t\in[t_a,t_d)\\ p_d+v_c(t-t_d)+\frac{1}{2}d(t-t_d)^2,&t\in[t_d,T]\end{cases} \]

Parameters

[in]	ctx	points to an instance of trapezoidal velocity trajectory
[in]	x	difference between current time and initial time

Returns

position output

a_trajtrap_vel()

double a_trajtrap_vel	(	a_trajtrap const *	ctx,
		double	x )

calculate velocity for trapezoidal velocity trajectory

\[ \dot{p}(t)=\begin{cases}v_0+at,&t\in[0,t_a)\\v_c,&t\in[t_a,t_d)\\v_c+d(t-t_d),&t\in[t_d,T]\end{cases} \]

Parameters

[in]	ctx	points to an instance of trapezoidal velocity trajectory
[in]	x	difference between current time and initial time

Returns

velocity output