Low-cost　microelectromechanical　inertial　measurement　units　(MIMUs)　are　widely　used　in　various　fields　such　as　healthcare,　sports,　industry,　and　indoor　navigation.　However,　rapid　and　robust　heading　alignment　under　dynamic　conditions　remains　a　challenge.　A　novel　heading　alignment　method　of　low-cost　MIMU　using　position　loci　for　indoor　navigation　is　proposed.　Aiming　at　the　state-of-the-art　acceleration-based　optimization-based　alignment　(ABOBA)　method＇s　vulnerability　to　dynamic　disturbances,　the　observation　equation　of　the　initial　attitude　matrix　is　established　based　on　velocity　and　position　loci.　On　this　basis,　Davenport＇s　q-method　is　adopted　to　estimate　the　initial　attitude.　Field　tests　are　conducted　under　dynamic　conditions　with　about　0.7　m/s　to　verify　the　performance　of　the　proposed　method　and　the　effect　of　the　referenced　position　error　on　the　heading　alignment　is　explored.　Results　show　that　the　initial　heading　error　could　be　quickly　converged　to　less　than　4　degrees　within　5　s,　which　is　far　better　than　the　state-of-the-art　method　and　is　of　significant　merits　to　solve　the　challenge　of　in-motion　accurate　heading　alignment　under　weak　acceleration　conditions.