Sparrow　Search　Algorithm(SSA)　is　a　new　swarm　intelligent　optimization　algorithm　that　simulates　the　foraging　behavior　and　anti-predation　behavior　of　sparrows.　It　has　the　advantages　of　simple　calculation,　fast　convergence,　and　few　parameters　that　need　to　be　adjusted.　However,　in　the　application　of　mobile　robot　path　planning,　there　is　insufficient　population　diversity　in　the　later　iterations　and　it＇s　easy　to　fall　into　a　locally　optimal　solution.　In　order　to　solve　these　problems　of　the　sparrow　search　algorithm　in　the　path　planning　of　mobile　robots,　an　improved　sparrow　search　algorithm　is　proposed.　First,　the　algorithm　uses　opposition-based　learning　to　optimize　the　initial　population　of　sparrows,　improves　the　quality　of　the　initial　solution,　and　enhances　the　local　search　ability　of　the　algorithm.　Second,　it　mixes　the　Metropolis　criterion　in　the　simulated　annealing　algorithm　and　makes　the　algorithm　able　to　accept　the　new　solution　by　judging　whether　to　accept　the　new　solution,　so　as　Jumping　out　of　the　local　optimum　and　strengthen　the　global　search　capability.　Finally,　the　performance　of　the　proposed　improved　algorithm　is　verified　on　2D　grid　maps　of　different　specifications　built　on　the　MATLAB　platform.　The　simulation　results　show　that　the　improved　SSA　has　better　optimization　performance　than　the　SSA　and　Particle　Swarms　Optimization　and　other　traditional　intelligent　algorithms,　and　can　effectively　jump　out　of　the　local　optimum　and　also　plan　a　safe　and　feasible　movement　with　the　best　cost　and　satisfying　constraints　in　a　stable　and　efficient　manner.　©　2021　IEEE.