Autonomous driving paper index
Developing and preliminary testing of temporal occlusion decision scenarios in dynamic contexts: a three-step method
One-line summary
Temporal occlusion is a method to segment recorded or created decision scenarios into constituent parts, to enable studying judgment and decision-making.
Engineering notes
Key topics: autonomous driving. See the paper for implementation details and experimental results.
Chinese explanation / 中文解读
中文解读待补充:本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。
Original abstract
Temporal occlusion is a method to segment recorded or created decision scenarios into constituent parts, to enable studying judgment and decision-making. There is increasing interest in using the method in complex domains such as police education, which calls for guiding principles in constructing and evaluating scenarios. We suggest principles for a three-step method to produce temporally occluded scenarios, comprising (1) extraction : how to identify and systematically select relevant scenarios, (2) production : how to develop materials, and (3) testing : iterative pilot testing of the materials. Further, we exemplify the method by producing and testing scenarios relevant for police work. In the extraction step, potential scenario descriptions and relevant research questions and outcome variables were decided upon after a discussion with domain experts. After evaluation, a subset of scenarios was produced as videos and temporally occluded, to allow for stopping scenarios at key points and asking for predictive judgments and plausible courses of action. In the testing step, materials were iteratively pilot tested, resulting in the finalized set of temporally occluded materials. This three-step process and the preliminary results provide initial guidance contributing to the standardization of temporally occluded scenario development that can help researchers gain a better understanding of anticipation and decision-making in different domains. Applications for this three-step method and future research possibilities are discussed.
Links and sources
Need this topic turned into a technical roadmap?
Full Self Driving can prepare a custom autonomous driving literature review, code map, dataset map, and B2B technology assessment.
Request B2B research
Comments