solsticestudioai/hyperion-drive-pack

Hyperion Drive AV安全智能包（示例）是一个合成的自动驾驶车辆遥测数据集，专为ADAS验证、传感器融合模型训练和边缘案例安全研究设计。每条记录代表一个完整的驾驶事件链，从环境触发（如黑冰、眩光、CAN总线故障）到感知或控制故障的连锁反应，包含高频CAN总线信号、故障类别标签、因果链注释和安全关键结果。数据集由SolsticeAI构建，作为更大商业包的免费样本，100%合成，不含真实车辆日志或可识别用户数据，适用于模型训练、监管评估和公共基准测试。数据集包含10,000个驾驶事件链，平衡分布在4个安全关键性等级和3个故障类别中，涵盖多种车辆模型、驾驶模式和环境混乱配置文件。

The Hyperion Drive AV Safety Intelligence Pack (Sample) is a synthetic autonomous-vehicle telemetry dataset designed for ADAS validation, sensor-fusion model training, and edge-case safety research. Each row represents a complete driving event chain — from environmental triggers (black ice, glare, CAN bus glitch) through cascading perception or control failures — with high-frequency CAN bus signals, failure-class labels, causal-chain annotations, and safety-critical outcomes. Built by SolsticeAI as a free sample of a larger commercial pack, it is 100% synthetic, containing no real vehicle logs or identifiable user data, making it safe for model training, regulator evaluation, and public benchmark work. The dataset includes 10,000 driving event chains, balanced across 4 safety-criticality tiers and 3 failure classes, covering various vehicle models, drive modes, and environmental chaos profiles.