Searcharxiv⌕ Search

arXiv subjects

Ruihai Ma

Publications and source records attributed to Ruihai Ma.

2 recordsLinked to original sources

ExtremeDrive: A Real-World Dataset and Benchmark for Critical Driving

Existing autonomous driving datasets and benchmarks provide extensive resources for routine driving, yet real-world data covering safety-critical scenarios and extreme environmental conditions remain limited. To address this gap, we present ExtremeDrive, a real-world dataset and benchmark covering emergency maneuvers, complex traffic interactions, and adverse conditions such as slippery or snow-covered roads, alongside routine driving. The dataset comprises 589 driving scenes curated from a larger pool of real-world driving recordings, with calibrated and aligned multisensor data and reviewed annotations of road structure and traffic participants. We further establish a benchmark to evaluate planned trajectories through simulated vehicle execution, measuring safety-related metrics and driving performance. Experiments at equal training set sizes show that incorporating critical driving data reduces trajectory prediction errors and improves mean critical-success rates in critical scenarios. ExtremeDrive provides a unified resource for learning and evaluating autonomous driving in safety-critical scenarios and extreme environmental conditions.

cs.RO↗

Quantum Secret Reconstruction

In addition to secret splitting, secret reconstruction is another important component of secret sharing. In this paper, the first quantum secret reconstruction protocol based on cluster states is proposed. Before the protocol, a classical secret is divided into multiple shares, which are distributed among shareholders via secret splitting. In the protocol, the dealer utilizes her secret to encrypt a private quantum state, and sends the encrypted state to a combiner chosen by her from the shareholders. With the help of other shareholders, the combiner utilizes the properties of cluster states to recover the privacy quantum state. It is shown that the proposed protocol is secure against several common attacks, including external and internal attacks. Compared with classical secret reconstruction protocols, this protocol not only achieves theoretical security of all shares, but also is more efficient due to reducing the distribution cost and computation cost. To demonstrate the feasibility of the protocol, a corresponding simulation quantum experiment is conducted on the IBM Q platform. Furthermore, in conjunction with quantum fingerprinting, it can be directly applied to achieve the task of multiple secrets sharing, because the classical shares can be reused in the proposed protocol.

quant-ph↗