Entrotaxis-Jump as a hybrid search algorithm for seeking an unknown emission source in a large-scale area with road network constraint

Abstract

In a sudden hazardous material leakage accident, the rapid and accurate localization of the leakage source can effectively reduce casualties and property losses. Utilizing the sensible robot to seek an unknown emission source has become a promising field, while most researches in this field fail to consider some intractable but practical factors, such as the large spatial scale of some search domains and the road network (that can obstruct robot’s maneuver). This paper proposes an efficient search algorithm, named as Entrotaxis-Jump, to seek an unknown emission source and obtain other source terms (e.g., source strength) in a large-scale (>0.1 km2) practical scene with road network constraints, such as a chemical cluster. The hybrid algorithm incorporates the Entrotaxis algorithm (a kind of cognitive search algorithm) with the intermittent search strategy, so it can utilize the triggering jump motion to alleviate the negative factors for search (road network constraints, expansive search domain, and turbulence effect). We select a chemical cluster in Shanghai, China as the typical research area and conduct a series of simulations in it to compare the performance of the Entrotaxis with the Entrotaxis-Jump under various release strength Q and wind speed V. The performance is reflected by the success rate (SR) and mean search time (MST), and we propose a skill score S to consider the two indexes synthetically. The results denote that the Entrotaxis-Jump outperforms Entrotaxis in all of our simulated scenarios, especially when the wind speed V > 2, in which case the SR of Entrotaxis drops sharply while the SR of the Entrotaxis-Jump witnesses a little decline but remains over 90%. The Entrotaxis-Jump algorithm proposed in this paper considers more practical factors, compared to previous researches, and is suitable and robust to utilize in real scenarios.