Health monitoring of controller area network in hybrid excavator based on the message response time
Dahui Gao, Qingfeng Wang · 2014
The work environment of the controller area network (CAN) in hybrid excavator is very harsh. Health monitoring of the CAN is of great importance. This work proposes a method to assess the performance of CAN and identify the faults base on the statistical characteristics of the message response time. This method can be easily embedded into a node to on-line monitor the health condition of CAN without using a special device. Firstly, we use the message response time to generate a feature to assess the health condition of CAN by comparing the difference between feature's probability density functions under different conditions base on the Bhattacharyya distance. Then, we propose a method to detect the fault on CAN and determine the fault type and fault duration. Finally, experiments are carried out on a hybrid excavator to verify the effectiveness of the proposed method, and the results show that the proposed method is applicable for the health monitoring of CAN.