Autonomous driving paper index
A multi-objective optimization framework for sustainable automotive interior design integrating enhanced triple bottom line, fuzzy decision-making, and BO-BiLSTM-driven NSGA-II
One-line summary
Reconciling aesthetic appeal with lifecycle sustainability remains a critical challenge in product design, particularly when subjective user perceptions and multidimensional sustainability targets must be jointly considered at the conceptual design stage.
Engineering notes
Key topics: autonomous driving, deployment, perception, prediction. 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
Reconciling aesthetic appeal with lifecycle sustainability remains a critical challenge in product design, particularly when subjective user perceptions and multidimensional sustainability targets must be jointly considered at the conceptual design stage. This study proposes an integrated decision-support framework for aesthetic–sustainability-oriented automotive interior design. An enhanced triple bottom line (TBL) model, augmented with Life Cycle Thinking and Doughnut Economics boundaries, is first introduced to construct a lifecycle-aware and boundary-constrained sustainability evaluation basis. User aesthetic imagery is captured through Kansei Engineering and prioritised using Interval Type-2 Fuzzy Hesitant Analytic Hierarchy Process, while sustainability indicators are identified and ranked within the enhanced TBL framework. Interval 2-Tuple q-Rung Orthopair Fuzzy Quality Function Deployment is then used to translate the prioritised aesthetic and sustainability requirements into quantifiable design features. A Bayesian Optimization-tuned Bidirectional Long Short-Term Memory surrogate model is constructed to predict four aesthetic–sustainability outputs, namely aesthetic, economic, environmental, and social performance. The trained surrogate is embedded into Non-dominated Sorting Genetic Algorithm II to generate Pareto-optimal solutions within the Doughnut-defined safe and just operating space. SHapley Additive exPlanations (SHAP) are further employed to interpret the feature contribution patterns across the four output dimensions. The automotive interior design case demonstrates that the proposed framework can generate balanced Pareto-optimal solutions with reliable surrogate prediction performance. SHAP results indicate that D1 is the dominant driver of aesthetic and economic performance, D12 contributes most strongly to environmental performance, and D5 has the greatest influence on social performance, while D7 shows cross-dimensional importance. The framework provides a transparent and data-driven paradigm for integrating aesthetic imagery, market feasibility, environmental responsibility, and social user experience in early-stage product development.
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