diff --git a/src/modules/mc_att_control/AttitudeControl/AttitudeControl.cpp b/src/modules/mc_att_control/AttitudeControl/AttitudeControl.cpp
index 4d48759ec6..7a348117f7 100644
--- a/src/modules/mc_att_control/AttitudeControl/AttitudeControl.cpp
+++ b/src/modules/mc_att_control/AttitudeControl/AttitudeControl.cpp
@@ -53,7 +53,7 @@ void AttitudeControl::setProportionalGain(const matrix::Vector3f &proportional_g
 	_proportional_gain(2) = roll_pitch_gain;
 }
 
-matrix::Vector3f AttitudeControl::update(matrix::Quatf q, matrix::Quatf qd, float yawspeed_feedforward)
+matrix::Vector3f AttitudeControl::update(matrix::Quatf q, matrix::Quatf qd, const float yawspeed_feedforward)
 {
 	// ensure input quaternions are exactly normalized because acosf(1.00001) == NaN
 	q.normalize();
diff --git a/src/modules/mc_att_control/AttitudeControl/AttitudeControlTest.cpp b/src/modules/mc_att_control/AttitudeControl/AttitudeControlTest.cpp
index ad07cb9d86..ab86085509 100644
--- a/src/modules/mc_att_control/AttitudeControl/AttitudeControlTest.cpp
+++ b/src/modules/mc_att_control/AttitudeControl/AttitudeControlTest.cpp
@@ -33,12 +33,106 @@
 
 #include <gtest/gtest.h>
 #include <AttitudeControl.hpp>
+#include <mathlib/math/Functions.hpp>
 
 using namespace matrix;
 
 TEST(AttitudeControlTest, AllZeroCase)
 {
 	AttitudeControl attitude_control;
-	matrix::Vector3f rate_setpoint = attitude_control.update(Quatf(), Quatf(), 0.f);
+	Vector3f rate_setpoint = attitude_control.update(Quatf(), Quatf(), 0.f);
 	EXPECT_EQ(rate_setpoint, Vector3f());
 }
+
+class AttitudeControlConvergenceTest : public ::testing::Test
+{
+public:
+	AttitudeControlConvergenceTest()
+	{
+		_attitude_control.setProportionalGain(Vector3f(.5f, .6f, .3f));
+		_attitude_control.setRateLimit(Vector3f(100, 100, 100));
+	}
+
+	void checkConvergence()
+	{
+		int i; // need function scope to check how many steps
+		Vector3f rate_setpoint(1000, 1000, 1000);
+		printf("Iterations: ");
+
+		for (i = 100; i > 0; i--) {
+			printf("%d ", i);
+			// run attitude control to get rate setpoints
+			const Vector3f rate_setpoint_new = _attitude_control.update(_quat_state, _quat_goal, 0.f);
+
+			// expect the error and hence also the output to get smaller with each iteration
+			if (rate_setpoint_new.norm() >= rate_setpoint.norm()) {
+				break;
+			}
+
+			rate_setpoint = rate_setpoint_new;
+			// rotate the simulated state quaternion according to the rate setpoint
+			_quat_state = _quat_state * Quatf(AxisAnglef(rate_setpoint));
+		}
+
+		printf("\n");
+
+		// it shouldn't have taken longer than an iteration timeout to converge
+		EXPECT_GT(i, 0);
+		// we need to have reached the goal attitude
+		EXPECT_EQ(antipodal(_quat_state), antipodal(_quat_goal));
+	}
+
+	Quatf antipodal(const Quatf q)
+	{
+		return q * math::signNoZero(q(0));
+	}
+
+	AttitudeControl _attitude_control;
+	Quatf _quat_state;
+	Quatf _quat_goal;
+};
+
+TEST_F(AttitudeControlConvergenceTest, AttitudeControlConvergenceUnit)
+{
+	_quat_state = Quatf();
+	checkConvergence();
+}
+
+TEST_F(AttitudeControlConvergenceTest, AttitudeControlConvergenceRoll180)
+{
+	_quat_state = Quatf(0, 1, 0, 0);
+	checkConvergence();
+}
+
+TEST_F(AttitudeControlConvergenceTest, AttitudeControlConvergencePitch180)
+{
+	_quat_state = Quatf(0, 0, 1, 0);
+	checkConvergence();
+}
+
+TEST_F(AttitudeControlConvergenceTest, AttitudeControlConvergenceYaw180)
+{
+	_quat_state = Quatf(0, 0, 0, 1);
+	checkConvergence();
+}
+
+TEST_F(AttitudeControlConvergenceTest, AttitudeControlConvergenceRandom)
+{
+	const Quatf QRandom[] = {
+		Quatf(0.698f, 0.024f, -0.681f, -0.220f),
+		Quatf(-0.820f, -0.313f, 0.225f, -0.423f),
+		Quatf(0.599f, -0.172f, 0.755f, -0.204f),
+		Quatf(0.216f, -0.662f, 0.290f, -0.656f)
+	};
+
+	for (int i = 0; i < 4; i++) {
+		for (int j = 0; j < 4; j++) {
+			printf("Random combination: %d %d\n", i, j);
+			_quat_state = QRandom[i];
+			_quat_goal = QRandom[j];
+			_quat_state.normalize();
+			_quat_goal.normalize();
+			checkConvergence();
+		}
+	}
+}