Robotically steering needles and other minimally invasive instruments towards diseased tissue for diagnosis or treatment has the potential to minimize targeting errors and improve patient care. In such procedures, it is often not desired to have fully automated control of the instrument. We develop a system which uses shared control of these instruments where both the physician and the automated controller provide steering inputs. Optimal navigation cues are provided through a combination of haptic (vibratory) and visual feedback to the operator who controls a slave robot to steer the needle. This allows the automated controller to provide feedback to the physician in order to reach the intended goal, while keeping the physician in control of the system. Experimental results show that receiving feedback from the control algorithm improves the targeting accuracy by a factor of 9 with respect to manual insertions. Additional project information is also available at the H-Haptics website.