Research On Path Planning Of Underwater Cleaning Robot

The land-based industrial aquaculture mode has been widely promoted and developed in China,and intelligent mechanization equipment plays an important role in the cultivation task.In aquaculture,fish manure,bait and microbial flora will be attached to the bottom and wall of the tank,which will affect the water quality and lead to disease occurrence,and it needs to be removed regularly.Traditional cleaning method needs manual hand-held board brush or dragging water gun machine around the brush,which is tedious and repetitive,and the bottom of the pool is easy to slide,and it is against the concept of energy conservation and environmental protection.The object of this paper-fish tank cleaning robot is mainly used to clean the bottom pollutants of fish tank in the aquaculture workshop,which solves the problems existing in manual cleaning,and the cleaning efficiency is obviously improved.In this paper,the author studies the mapping and positioning effect of robot in indoor fish tank environment and the ability of robot to clean the full coverage path according to the requirement of robot to realize high efficiency cleaning.This paper mainly studies the construction,location and full coverage path planning technology.The specific content of the paper includes the following aspects:(1)The research on the mapping and positioning of the fish tank cleaning robot is carried out.Firstly,the slam system model is established for the fish tank cleaning robot,and the environment features are extracted by sensors.Then three methods of building maps are introduced and compared.The map construction function based on gmapping algorithm is selected.Then,the research on location is carried out.Firstly,the shortcomings of the odometer estimation and the errors brought by the experiment are analyzed.Then,an Adaptive Monte Carlo localization algorithm is proposed to obtain the optimal estimation of the robot position and posture in the map.Finally,the relevant simulation experiments show that the algorithm is effective.(2)The path planning of fish tank cleaning robot is studied.Firstly,the concept of path planning algorithm is introduced in detail.Based on the problem of path planning,the modeling method in two-dimensional space is compared,and the grid chart method suitable for fish tank environment is selected.From the point of view of the design requirements of brush cleaning robot,it is necessary to traverse the covering work in the environment space.Then,the algorithm of full coverage path planning is introduced.The algorithm and performance evaluation index are mathematically explained.Through comparing the random,round-trip and internal spiral covering cleaning modes,the grid based full coverage path planning algorithm is proposed.Finally,the feasibility of the algorithm is verified by the design of simulation experiment.(3)The design and experimental test of the prototype of the fish tank cleaning robot are presented.Firstly,the overall design idea of the prototype is introduced,and the main control modules of the robot are STM32 controller and composite sensor.Then,through setting up the comparative experiment of different factors,the positioning performance of fish tank cleaning robot is analyzed.The robot is driven to complete the full coverage according to the internal spiral path planning algorithm Cover cleaning task.The experimental data and results show that the robot can walk stably,its walking speed can reach 10 m/min,the cleaning speed is about 1 min/m~2,and the working time is about 1h,which can reach the cleaning index of high coverage and low repetition rate.The robot can build a map with high precision in unknown environment and can complete the effective positioning of the robot.Under the condition of basic test group,the average positioning distance error is 9.02 cm,the average positioning angle error is4.6°,and it has good positioning and navigation ability;In the initial prototype test,regardless of the influence of brush plate shape and material,the fish pond cleaning coverage rate can reach 85%and the repetition rate is less than 10%,which has high efficiency cleaning ability.The research results are of reference value to the research of positioning navigation system and full coverage path planning in indoor fish tank environment,and provide the basic research direction for land-based industrial aquaculture cleaning equipment.