Various　sources　of　error　can　lead　to　the　position　accuracy　of　the　robot　being　orders　of　magnitude　worse　than　its　repeatability.　For　the　accuracy　of　drilling　in　the　aviation　field,　high-precision　assembly,　and　other　areas　depending　on　the　industrial　robot＇s　absolute　positioning　accuracy,　it　is　essential　to　improve　the　accuracy　of　absolute　positioning　through　calibration.　In　this　paper,　an　error　model　of　the　robot　considering　both　constant　and　high-order　joint-dependent　kinematic　errors　is　established,　and　the　model　is　modified　by　the　Hermite　polynomial,　thereby　mitigating　the　occurrence　of　the　Runge　phenomenon.　To　identify　high-order　joint-dependent　kinematic　errors,　a　robot　calibration　method　based　on　the　back-propagation　neural　network　(BP)　optimized　by　the　sparrow　search　algorithm　(SSA-BP)　is　proposed,　which　optimizes　the　uncertainty　of　weights　and　thresholds　in　the　BP　algorithm.　Experiments　on　an　EFORT　ECR5　robot　were　implemented　to　validate　the　efficiency　of　the　proposed　method.　The　positioning　error　is　reduced　from　3.1704　mm　to　0.2798　mm,　and　the　error　decrease　rate　reaches　42.92%　(compared　with　BP　calibration)　and　21.09%　(compared　with　particle　swarm　optimization　back-propagation　calibration).　With　the　new　calibration　method　using　SSA-BP,　robot　positioning　errors　can　be　effectively　compensated　for,　and　the　robot　positioning　accuracy　can　be　improved　significantly.