| The proportion of cables in complex mechatronic products is increasing,which makes cables a key factor affecting product assembly,performance and quality.The internal space of complex mechatronic products is narrow and there are many cables and connectors,which poses a challenge to the formulation of the cable assembly process.At present,whether the traditional method uses physical prototypes to simulate the assembly of cables,or the computer-aided technology is used to simulate the assembly of cables,most of the obtained cable assembly process files are not related to the actual physical assembly situation,and there is no better human-computer interaction issues.In order to improve the quality and efficiency of cable assembly,this paper studies several key technologies of complex mechatronic products cable assembly,mainly including the following aspects:（1）In order to increase the efficiency of solving cable assembly sequence,a method for solving cable assembly sequence based on improved particle swarm algorithm is proposed.First,in order to formulate a suitable objective function,the cable assembly sequence model is constructed and used as the evaluation index of the objective function.Then,in order to adapt to the discrete characteristics of cable assembly sequence planning,the basic particle swarm optimization algorithm is discretized.At the same time,for stopping the global optimal solution Gbest from falling into the local optimal,an improved algorithm is proposed to mutate Gbest and perturb the update method of particle velocity by elite particles.Finally,the algorithm is tested with three Benchmark functions,and the algorithm is applied to a model with 17groups of cables for verification.（2）In view of the problems that the cable model cannot express the physical characteristics of the cable well,and the intelligent algorithm cannot be well adapted to the cable assembly path solution,etc,this paper proposes a method for solving the cable assembly path based on an improved rapidly-exploring random tree algorithm.This method first builds a cable physical property model based on the Cosserat elastic rod theory,and then proposes three methods to improve the fast search random tree algorithm:（1）On the premise of facilitating the actual cable assembly,a cable assembly path planning strategy based on the artificial potential field method is proposed to connect the cable assembly path to the target configuration of the cable.（2）In order to prevent the cable from colliding and interfering with the components due to its own gravity,a node expansion strategy considering the optimal height of the cable from the ground is proposed.（3）To find out a relatively smooth path,a path optimization method based on a non-uniform cubic B-spline curve is proposed.Finally,the algorithm in this chapter is verified by experimental simulation and example analysis.（3）Aiming at the problems that the current ideal assembly scheme can not meet the actual needs and so on,a dynamic construction method of cable assembly process model of complex mechatronic products based on digital twin is proposed.Use SLAM to map the component features during cable assembly and cable spatial positioning,introduce RFS to describe SLAM problems,and use GM-PHD to realize map components prediction and update of feature information;simultaneously,SRTUKF and single feature strategy are used to estimate the pose of the cable assembly path,thereby constructing a digital twin of the cable assembly path to realize the dynamic construction of the cable assembly process model.Finally,taking the cable assembly process of an electrical cabinet model as an example,the reliability of the method is verified by experimental simulation.（4）This paper uses Microsoft Visual Studio 2010,Creo2.0 and its Creo/Toolkit software to develop a cable assembly process planning prototype system,and demonstrates its workflow in an example. |
