small improvement to sensor fusion
1) there was a typo when computing the system covariance
a term in dT^3 was ommitted; the impact was was very limited
because of how small this term is.
2) initialize the system covariance matrix with non-zero
values for the gyro-bias part. this improves the initial
bias estimation speed significantly.
3) added comments here and there
Change-Id: I4328c9cca73e089889d5e74b9fda99d7831762dc
diff --git a/services/sensorservice/Fusion.cpp b/services/sensorservice/Fusion.cpp
index b724ce2..2b6b793 100644
--- a/services/sensorservice/Fusion.cpp
+++ b/services/sensorservice/Fusion.cpp
@@ -201,15 +201,15 @@
// q11 = su^2.dt
//
- // variance of integrated output at 1/dT Hz
- // (random drift)
- const float q00 = gyroVAR * dT;
+ const float dT2 = dT*dT;
+ const float dT3 = dT2*dT;
+
+ // variance of integrated output at 1/dT Hz (random drift)
+ const float q00 = gyroVAR * dT + 0.33333f * biasVAR * dT3;
// variance of drift rate ramp
const float q11 = biasVAR * dT;
-
- const float u = q11 / dT;
- const float q10 = 0.5f*u*dT*dT;
+ const float q10 = 0.5f * biasVAR * dT2;
const float q01 = q10;
GQGt[0][0] = q00; // rad^2
@@ -220,6 +220,22 @@
// initial covariance: Var{ x(t0) }
// TODO: initialize P correctly
P = 0;
+
+ // it is unclear how to set the initial covariance. It does affect
+ // how quickly the fusion converges. Experimentally it would take
+ // about 10 seconds at 200 Hz to estimate the gyro-drift with an
+ // initial covariance of 0, and about a second with an initial covariance
+ // of about 1 deg/s.
+ const float covv = 0;
+ const float covu = 0.5f * (float(M_PI) / 180);
+ mat33_t& Pv = P[0][0];
+ Pv[0][0] = covv;
+ Pv[1][1] = covv;
+ Pv[2][2] = covv;
+ mat33_t& Pu = P[1][1];
+ Pu[0][0] = covu;
+ Pu[1][1] = covu;
+ Pu[2][2] = covu;
}
bool Fusion::hasEstimate() const {
@@ -357,6 +373,11 @@
mat34_t Fusion::getF(const vec4_t& q) {
mat34_t F;
+
+ // This is used to compute the derivative of q
+ // F = | [q.xyz]x |
+ // | -q.xyz |
+
F[0].x = q.w; F[1].x =-q.z; F[2].x = q.y;
F[0].y = q.z; F[1].y = q.w; F[2].y =-q.x;
F[0].z =-q.y; F[1].z = q.x; F[2].z = q.w;
@@ -413,7 +434,12 @@
K[1] = transpose(P[1][0])*LtSi;
// update...
- // P -= K*H*P;
+ // P = (I-K*H) * P
+ // P -= K*H*P
+ // | K0 | * | L 0 | * P = | K0*L 0 | * | P00 P10 | = | K0*L*P00 K0*L*P10 |
+ // | K1 | | K1*L 0 | | P01 P11 | | K1*L*P00 K1*L*P10 |
+ // Note: the Joseph form is numerically more stable and given by:
+ // P = (I-KH) * P * (I-KH)' + K*R*R'
const mat33_t K0L(K[0] * L);
const mat33_t K1L(K[1] * L);
P[0][0] -= K0L*P[0][0];