The　rapid　development　of　smart　sensor　and　artificial　intelligence　(AI)　technologies　has　created　a　huge　demand　for　multifunctional　sensors　to　address　energy　harvesting　and　accurate　sensing　in　complex　environments.　However,　it　is　still　a　major　challenge　to　fabricate　self-powered　and　functional　sensors　with　excellent　mechanical　strength.　Herein,　this　study　utilized　vacuum-assisted　self-assembly　technology　to　combine　conductive　Nb4C3Tx　with　bacterial　cellulose　(BC)　to　construct　Nb4C3Tx-BC　(NBC)　composites　with　a　three-dimensional　network.　Benefiting　from　abundant　hydrogen　bonds,　NBC　composites　demonstrate　significant　fracture　stress　(44.8　MPa)　and　satisfactory　Young＇s　modulus　(1.91　GPa).　While　integrating　the　excellent　conductivity　of　Nb4C3Tx,　NBC　composites　achieve　a　conductivity　as　high　as　7.1　S　m−1.　The　Seebeck　coefficient　of　the　thermoelectric　nanogenerators　made　with　NBC　composites　can　reach　−7.83　μV　K−1.　At　the　same　time,　the　triboelectric　nanogenerators　based　on　NBC　composites　exhibit　the　output　voltage　and　power　density　of　64　V　and　2.85　mW　m−2,　respectively.　Based　on　these　characteristics,　multimodal　self-powered　sensors　have　been　developed　toward　a　material/temperature　recognition　system.　For　example,　water　temperature　detection　mechanical　claws,　self-powered　pianos,　and　breathing　detection　sensors　were　developed　based　on　temperature　induced　self-powered　sensing.　Furthermore,　by　combining　with　multilayer　perceptron　neural　networks,　a　self-powered　writing　tablet　and　a　multimodal　material/temperature　recognition　system　were　developed.　The　NBC　composites　based　multimodal　self-powered　sensors　demonstrates　simultaneous　detection　of　dynamic　stimuli,　temperature　variations,　and　surface　material　properties,　making　it　suitable　for　advanced　applications　in　smart　robotics.　©　2025　Elsevier　B.V.