The　invention　discloses　a　six-degree-of-freedom　robot　path　planning　method　based　on　an　end　motion　error,　　and　specifically,　　a　method　of　accurately　planning　a　continuous　path　for　a　robot　end　portion　by　combining　a　screw　theory,　　a　cubic　spline　interpolation　algorithm　and　a　particle　swarm　optimization　algorithm.　The　method　comprises　:　　establishing　a　robot　forward-inverse　kinematics　model　based　on　the　screw　theory　to　simplify　a　computation　process;　　adopting,　　in　a　joint　space,　　the　cubic　spline　interpolation　algorithm　to　ensure　a　smooth　motion;　　and　last,　　utilizing　a　count　of　key　points　as　a　variable　to　control　an　end　tracking　error　within　a　required　range,　　utilizing　respective　time　intervals　as　a　design　variable,　　and　maximum　angular　velocities,　　maximum　angular　accelerations　and　maximum　angular　jerks　of　respective　joints　as　constraints,　　and　optimizing　a　path　by　using　the　minimum　tracking　error　as　an　optimization　target.　As　a　result,　　the　embodiment　is　employed　to　obtain,　　using　high　planning　efficiency,　　a　planned　path　having　a　low　tracing　error　and　producing　a　smooth　motion.