Compensating for Operational Uncertainty in Man–Machine Systems: A Case Study on Intelligent Vehicle Parking Assist System

Abstract

The successful operation of man–machine systems requires consistent human operation and reliable machine performance. Machine reliability has received numerous improvements, whereas human-related operational uncertainty is an area of increasing research interest. Most studies and formal documentation only provide suggestions for alleviating human uncertainty instead of providing specific methods to ensure operation accuracy in real-time. This paper presents a general framework for a reliable system that compensates for human-operating uncertainty during operation. This system learns the response of the user, constructs the user’s behavior pattern, and then creates compensated instructions to ensure the completion of the desired tasks, thus improving the reliability of the man–machine system. The proposed framework is applied to the development of an intelligent vehicle parking assist system. Existing parking assist systems do not account for driver error, nor do they consider realistic urban parking spaces with obstacles. The proposed system computes a theoretical path once a parking space is identified. Audio commands are then sent to the driver with real-time compensation to minimize deviations from the path. When an operation is too far away from the desired path to be compensated, a new set of instructions is computed based on the collected uncertainty. Tests with various real-world urban parking scenarios indicated that there is a possibility to park a vehicle with a space that is as small as 1.07 times the vehicle length with up to 30% uncertainty. Results also show that the compensation scheme allows diverse operators to reliably achieve a desired goal.