The　trajectory　planning　with　obstacle　avoidance　is　a　critical　task　for　redundant　manipulators　in　the　complex　operating　environment.　This　paper　proposes　a　time-minimum　trajectory　optimization　technique　to　obtain　a　collision-free　and　jerk-continuous　trajectory　for　a　4-dof　redundant　manipulator　considering　both　kinematic　and　spatial　constraints.　In　the　optimization,　the　whole　trajectory　is　divided　into　several　sub-trajectories　to　ensure　the　reliability　of　the　technique.　The　screw-based　inverse　kinematics　(IK)　model　is　used　to　obtain　the　angular　displacements　of　joints　corresponding　to　the　end-effector　final　positions　in　each　sub-trajectory.　The　spatial　constraint　is　determined　by　assuming　robotic　links　using　cylinders　with　different　bottom　radius,　and　enveloping　obstacles　using　spheres　under　different　sizes.　The　quintic　polynomial　curve　is　utilized　to　interpolate　in　joint　space　to　guarantee　the　continuous　of　the　angular　jerk　of　each　joint.　Finally,　PSO　algorithm　is　used　to　search　for　the　optimal　trajectory　of　the　robot　by　taking　the　whole　motion　time　as　the　optimization　objective.　Two　cases　are　taken　to　validate　the　effectiveness　of　the　proposed　technique.　The　simulation　results　in　MATLAB　show　that　the　robotic　motion　based　on　the　optimal　trajectory　are　safe,　stable　and　efficient.