Laser radar water quality inspection system for recirculating aquaculture workshop based on multi-objective genetic algorithm

With the expansion of the scale of the aquaculture workshop, the types and quantities of fish in the aquaculture pond have increased significantly, which is prone to water pollution or water quality changes, resulting in production safety accidents in the aquaculture workshop. The use of water quality inspection system for regular inspection of aquaculture ponds can not only reduce the labor intensity of aquaculture workers, but also detect abnormal water quality in aquaculture ponds in time and reduce economic losses. Due to the high humidity and poor lighting conditions in the recirculating aquaculture workshop, and the existence of a large number of aquaculture facilities and supporting aquaculture equipment, if a single target path planning method is adopted, it cannot meet the operational requirements of the water quality inspection system. To solve the above problems, this paper proposes a water quality inspection system based on three-dimensional lidar positioning technology. At the same time, a multi-objective genetic algorithm is used to solve the optimal inspection path of the robot considering five main factors, including safety, distance, stability, travel duration and collision-free path. And the feasibility of the water quality inspection system is verified by experiments. In the process of multi-objective genetic algorithm calculation, the number of path filling is first initialized; then, multiple objective functions are calculated to measure the fitness of individual paths, and the path satisfying the fitness criterion is selected as the optimal path of the water quality inspection system. The best individual is selected from the population with its probability by the tournament selection method, and then the two chromosomes are exchanged. The new offspring are generated by cyclic crossover, and then the adaptive bit mutation is performed to randomly exchange the input bit string. After executing the genetic operator, the fitness of the path is calculated and the fitness standard is checked. Repeat the above process until the specified conditions are met. Finally, the optimal path suitable for the movement of the water quality inspection system is selected. The experimental results show that the minimum lateral deviation of robot navigation is 4.5 cm, the maximum deviation is 7.2 cm, the minimum longitudinal deviation is 3.6 cm, the maximum deviation is 6.3 cm, the minimum heading deviation is 4.1°, and the maximum deviation is 6.9°, which meets the navigation and positioning requirements of the breeding workshop. The detected water temperature, dissolved oxygen, and pH were completely consistent with the manual detection results, indicating that the water quality inspection system was stable and could replace the manual detection of water quality in the aquaculture workshop. The simulation results based on multi-objective genetic algorithm show that compared with MOVNS and RCHF methods, the travel time of this algorithm is reduced by 31% and 42% respectively under the same driving path, and the obstacle collision rate is reduced by 15% and 31% respectively under the same path and number of obstacles. When selecting the path with the minimum time complexity in path planning, the performance of the proposed algorithm is more efficient and smoother. The research results can provide reference for the development of water quality inspection system in aquaculture workshop.