The　task　allocation　problem　of　unmanned　delivery　vehicles　(UDVs)　is　critical　and　has　attracted　much　attention　in　the　field　of　intelligent　delivery.　However,　most　current　studies　are　limited　to　applications　since　they　simply　regard　all　UDVs　to　be　dispatched　as　identical.　In　this　paper,　a　cooperative　task　allocation　method　is　proposed　for　dispatching　heterogeneous　UDVs　to　complete　delivery　tasks.　First,　the　cooperative　task　allocation　problem　is　formulated　as　a　multi-objective　optimization　problem　subject　to　constraints,　where　two　conflicting　objectives　are　total　distance　and　differential　workload　balance.　The　differential　workload　balance　objective　is　to　minimize　the　highest　difference　between　the　vehicle　ability　and　the　workload　of　an　individual　vehicle.　Second,　a　discrete　particle　swarm　optimization　algorithm　with　a　priority-guided　correction　strategy　is　proposed　to　solve　the　multi-objective　optimization　problem.　The　priority-guided　correction　strategy　guarantees　the　particle　search　in　a　feasible　region　and　improves　the　search　efficiency　by　utilizing　heterogeneous　vehicle　information.　Third,　a　multi-mutation　operator　local　search　strategy　is　embedded　to　avoid　particles　easily　trapping　in　local　optima　during　the　search　process.　The　strategy　enhances　the　local　search　ability　by　combining　three　mutation　operators,　the　interchange　mutation　operator,　the　insertion　mutation　operator,　and　the　exchange　mutation　operator.　Simulation　results　demonstrate　the　effectiveness　and　advantages　of　the　proposed　optimization　algorithm　for　solving　the　cooperative　task　allocation　problem　of　heterogeneous　unmanned　delivery　vehicle.　©　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.　2024.