pushquantum-2024 / Sensor Positioning

In the automotive industry, newly produced vehicles must be transported from the production line to the distribution area. Using automated-driving software, this task can be performed without a human driver if the environment is covered by infrastructure-based sensors. The positions and angles of these sensors must be optimized to ensure complete coverage while minimizing their number and associated costs. This optimization problem, modeled as a Max-Cover problem, is NP-hard and typically solved with heuristic methods. However, Quantum Computing offers potential for improved solutions. Initial tests will focus on simplified scenarios to evaluate feasibility, scalability, and cost savings, paving the way for scalable, cost-efficient applications in factory environments.

In the automotive industry, autonomous transport robots are increasingly used for moving newly manufactured vehicles. To ensure safe and reliable operation, comprehensive surveillance using infrastructure-based sensors like LiDAR is required. Given the high costs of these sensors, optimizing their placement to minimize the number of sensors while ensuring full coverage of the surveillance area is essential.

This use case focuses on solving the sensor placement problem using a Quadratic Unconstrained Binary Optimization (QUBO) model. Additionally, we explore the potential of quantum computing, particularly quantum annealers, to address the computational challenges of large-scale configurations.