The paper introduces MMAUD, a comprehensive multi-modal anti-UAV dataset designed to address the evolving challenges posed by small unmanned aerial vehicles (UAVs). MMAUD focuses on drone detection, UAV-type classification, and trajectory estimation, leveraging diverse sensory inputs including stereo vision, LIDAR arrays, RADARs, and audio arrays. The dataset offers high-fidelity aerial detection capabilities and provides accurate ground truth data generated by Leica, enhancing the credibility and reliability of the dataset. Key contributions include: 1. **Multi-modal Data Integration**: MMAUD combines visual, LIDAR, RADAR, and audio data to provide a rich and diverse dataset for advanced UAV detection techniques. 2. **High-Accuracy Ground Truth**: Utilizing Leica-generated ground truth, MMAUD sets new benchmarks in millimeter-level accuracy, a feature not available in previous datasets. 3. **Cost-effective and Adaptable**: The dataset and sensor configurations are cost-effective and highly adaptable, making it suitable for mobile and life-saving applications. 4. **Real-world Simulations**: The dataset incorporates ambient heavy machinery sounds to enhance its applicability in real-world scenarios, particularly during proximate vehicular operations. The paper also discusses the challenges and limitations, such as limited geographic coverage, sensor synchronization issues, and the variability of drone models. Despite these challenges, MMAUD remains a valuable resource for developing precise anti-UAV solutions. The dataset, along with its associated codes and designs, is available on GitHub.The paper introduces MMAUD, a comprehensive multi-modal anti-UAV dataset designed to address the evolving challenges posed by small unmanned aerial vehicles (UAVs). MMAUD focuses on drone detection, UAV-type classification, and trajectory estimation, leveraging diverse sensory inputs including stereo vision, LIDAR arrays, RADARs, and audio arrays. The dataset offers high-fidelity aerial detection capabilities and provides accurate ground truth data generated by Leica, enhancing the credibility and reliability of the dataset. Key contributions include: 1. **Multi-modal Data Integration**: MMAUD combines visual, LIDAR, RADAR, and audio data to provide a rich and diverse dataset for advanced UAV detection techniques. 2. **High-Accuracy Ground Truth**: Utilizing Leica-generated ground truth, MMAUD sets new benchmarks in millimeter-level accuracy, a feature not available in previous datasets. 3. **Cost-effective and Adaptable**: The dataset and sensor configurations are cost-effective and highly adaptable, making it suitable for mobile and life-saving applications. 4. **Real-world Simulations**: The dataset incorporates ambient heavy machinery sounds to enhance its applicability in real-world scenarios, particularly during proximate vehicular operations. The paper also discusses the challenges and limitations, such as limited geographic coverage, sensor synchronization issues, and the variability of drone models. Despite these challenges, MMAUD remains a valuable resource for developing precise anti-UAV solutions. The dataset, along with its associated codes and designs, is available on GitHub.