Object Detection for Autonomous Dozers
This work addresses safety and efficiency for autonomous construction vehicles, but it is incremental as it applies standard methods to a new domain.
The paper tackled object detection for autonomous dozers on construction sites by collecting data and benchmarking two existing models with various training strategies, achieving performance improvements but without specifying concrete numbers.
We introduce a new type of autonomous vehicle - an autonomous dozer that is expected to complete construction site tasks in an efficient, robust, and safe manner. To better handle the path planning for the dozer and ensure construction site safety, object detection plays one of the most critical components among perception tasks. In this work, we first collect the construction site data by driving around our dozers. Then we analyze the data thoroughly to understand its distribution. Finally, two well-known object detection models are trained, and their performances are benchmarked with a wide range of training strategies and hyperparameters.