Abstract: In some embodiments, techniques are provided for analyzing time series data to detect anomalies. In some embodiments, the time series data is processed using a machine learning model. In some embodiments, the machine learning model is trained in an unsupervised manner on large amounts of previous time series data, thus allowing highly accurate models to be created from novel data. In some embodiments, training of the machine learning model alternates between a fitting optimization and a trimming optimization to allow large amounts of training data that includes untagged anomalous records to be processed. Because a machine learning model is used, anomalies can be detected within complex systems, including but not limited to autonomous vehicles such as unmanned aerial vehicles. When anomalies are detected, commands can be transmitted to the monitored system (such as an autonomous vehicle) to respond to the anomaly.

Abstract: In some embodiments, techniques are provided for verifying operability of an automatic dependent surveillance-broadcast (ADS-B) receiver included in a first unmanned aerial vehicle (UAV), which includes receiving ADS-B data representative of ADS-B messages broadcast by traffic within a reception range of the ADS-B receiver during a first period of time, estimating a traffic environment for a service area spanning, at least in part, a first operating area of the first UAV during the first period of time, determining an expected observed traffic of the first UAV during the first period of time based on the estimated traffic environment, and verifying operability of the ADS-B receiver of the first UAV based on a comparison between the expected observed traffic of the first UAV and the traffic associated with the ADS-B data received by the ADS-B receiver of the first UAV.

Abstract: In some embodiments, techniques are provided for analyzing time series data to detect anomalies. In some embodiments, the time series data is processed using a machine learning model. In some embodiments, the machine learning model is trained in an unsupervised manner on large amounts of previous time series data, thus allowing highly accurate models to be created from novel data. In some embodiments, training of the machine learning model alternates between a fitting optimization and a trimming optimization to allow large amounts of training data that includes untagged anomalous records to be processed. Because a machine learning model is used, anomalies can be detected within complex systems, including but not limited to autonomous vehicles such as unmanned aerial vehicles. When anomalies are detected, commands can be transmitted to the monitored system (such as an autonomous vehicle) to respond to the anomaly.

Abstract: In some embodiments, techniques are provided for verifying operability of an automatic dependent surveillance-broadcast (ADS-B) receiver included in a first unmanned aerial vehicle (UAV), which includes receiving ADS-B data representative of ADS-B messages broadcast by traffic within a reception range of the ADS-B receiver during a first period of time, estimating a traffic environment for a service area spanning, at least in part, a first operating area of the first UAV during the first period of time, determining an expected observed traffic of the first UAV during the first period of time based on the estimated traffic environment, and verifying operability of the ADS-B receiver of the first UAV based on a comparison between the expected observed traffic of the first UAV and the traffic associated with the ADS-B data received by the ADS-B receiver of the first UAV.

Abstract: In some embodiments, techniques are provided for verifying operability of an automatic dependent surveillance-broadcast (ADS-B) receiver included in a first unmanned aerial vehicle (UAV), which includes receiving ADS-B data representative of ADS-B messages broadcast by traffic within a reception range of the ADS-B receiver during a first period of time, estimating a traffic environment for a service area spanning, at least in part, a first operating area of the first UAV during the first period of time, determining an expected observed traffic of the first UAV during the first period of time based on the estimated traffic environment, and verifying operability of the ADS-B receiver of the first UAV based on a comparison between the expected observed traffic of the first UAV and the traffic associated with the ADS-B data received by the ADS-B receiver of the first UAV.

Abstract: In some embodiments, techniques are provided for verifying operability of an automatic dependent surveillance-broadcast (ADS-B) receiver included in a first unmanned aerial vehicle (UAV), which includes receiving ADS-B data representative of ADS-B messages broadcast by traffic within a reception range of the ADS-B receiver during a first period of time, estimating a traffic environment for a service area spanning, at least in part, a first operating area of the first UAV during the first period of time, determining an expected observed traffic of the first UAV during the first period of time based on the estimated traffic environment, and verifying operability of the ADS-B receiver of the first UAV based on a comparison between the expected observed traffic of the first UAV and the traffic associated with the ADS-B data received by the ADS-B receiver of the first UAV.

Abstract: In some embodiments, techniques are provided for analyzing time series data to detect anomalies. In some embodiments, the time series data is processed using a machine learning model. In some embodiments, the machine learning model is trained in an unsupervised manner on large amounts of previous time series data, thus allowing highly accurate models to be created from novel data. In some embodiments, training of the machine learning model alternates between a fitting optimization and a trimming optimization to allow large amounts of training data that includes untagged anomalous records to be processed. Because a machine learning model is used, anomalies can be detected within complex systems, including but not limited to autonomous vehicles such as unmanned aerial vehicles. When anomalies are detected, commands can be transmitted to the monitored system (such as an autonomous vehicle) to respond to the anomaly.