{"paper":{"title":"Radioactive Source Seeking using Bayesian Optimisation with Movement Penalty","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Bayesian optimisation with a movement penalty and heteroscedastic Gaussian process localises radioactive sources while achieving sublinear regret.","cross_cats":["cs.SY","eess.SY"],"primary_cat":"physics.app-ph","authors_text":"Airlie Chapman, Jeremy M. C. Brown, Joshua Keene, Lysander Miller","submitted_at":"2026-05-14T15:16:33Z","abstract_excerpt":"The use of mobile robotics in radioactive source seeking has become an important part of modern radiation-safety practices, supporting timely mitigation of contamination risks and helping protect public health. However, measuring radiation is often time-consuming, rendering traditional gradient-based source-seeking methods less effective due to lower sample efficiency. This paper proposes a sample-efficient Bayesian-Optimisation source-seeking strategy that utilises a heteroscedastic Gaussian process surrogate to balance exploration and exploitation. Excessive inter-sample travel is discourage"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The strategy is shown to generate sublinear regret in the source-seeking task, while simulations demonstrate its effectiveness in localising radioactive sources.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the radiation intensity field can be adequately modeled by a heteroscedastic Gaussian process whose noise structure remains stable enough for the regret bound to hold in realistic environments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A Bayesian optimization method with heteroscedastic Gaussian process surrogate and movement switching cost achieves sublinear regret and effective source localization in simulations for radioactive source seeking.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Bayesian optimisation with a movement penalty and heteroscedastic Gaussian process localises radioactive sources while achieving sublinear regret.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"de9c771d564607a2a04e171d71ba5e8f1155c792a6e6a8f3889997884a0f7fab"},"source":{"id":"2605.14942","kind":"arxiv","version":1},"verdict":{"id":"773fece1-76a2-46d6-bcfa-b09d66af8c24","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T03:16:51.109530Z","strongest_claim":"The strategy is shown to generate sublinear regret in the source-seeking task, while simulations demonstrate its effectiveness in localising radioactive sources.","one_line_summary":"A Bayesian optimization method with heteroscedastic Gaussian process surrogate and movement switching cost achieves sublinear regret and effective source localization in simulations for radioactive source seeking.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the radiation intensity field can be adequately modeled by a heteroscedastic Gaussian process whose noise structure remains stable enough for the regret bound to hold in realistic environments.","pith_extraction_headline":"Bayesian optimisation with a movement penalty and heteroscedastic Gaussian process localises radioactive sources while achieving sublinear regret."},"references":{"count":19,"sample":[{"doi":"","year":2009,"title":"Extremum seeking with stochastic perturbations , author=. IEEE Trans. Autom. Control , volume=. 2009 , publisher=","work_id":"9e8dc2b1-eba2-4eb2-826f-02e8cb76123e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2000,"title":"Performance improvement and limitations in extremum seeking control , journal =","work_id":"4cd2c38a-a9ea-4ad5-b52c-82a58d9e830d","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Atanasov, Nikolay and Le Ny, Jerome and Michael, Nathan and Pappas, George J. , booktitle=. Stochastic source seeking in complex environments , year=","work_id":"7691bd80-4759-4908-94b8-20468e9cd206","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Radioactive capsule in","work_id":"e7eec177-046d-4e3d-a9d4-ce22e6cc2697","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Kakade and Matthias W","work_id":"605302b6-789a-4ed6-baed-2e838868029d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":19,"snapshot_sha256":"54436d35fddf4d997c47f399d916c81106d3f8cff4feee894ad22e8e3f768db2","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"2e4837d65944b945d627735721965a1e14b94735f13e9de7bb71dfd7e244d613"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}