Abstract: A system and method are disclosed for monitoring rides in a vehicle in which a driver of the vehicle picks up a rider at a pickup location and drives the rider to a drop-off destination. The system includes at least one sensor arranged in the vehicle and configured to capture sensor data during the rides, a transceiver configured to communicate with a personal electronic device of a driver of the vehicle, a non-volatile memory configured to store data; and a processor. The system captures sensor data during a ride, receives a ride identifier from the personal electronic device that uniquely identifies the ride, and stores the sensor data captured during the ride with the ride identifier as metadata.

Abstract: A cloud storage backend for managing data of in-vehicle sensing modules is disclosed, which is advantageous for use in the context of a shared vehicle service, such as a car rental service, an autonomous taxi service, or a ride sharing service. The in-vehicle sensing modules are configured to monitor a status of the vehicle and utilize appropriate algorithms, models, or thresholds to interpret sensor data and enrich the data with metadata and event detection. The cloud storage backend receives relevant sensor data, event data, or other metadata from the in-vehicle sensing modules and stores the data in a database which is made accessible by authorized third-parties.

Abstract: Methods for safe and supervised ride hailing experiences by a mobility service provider are disclosed. The methods advantageously utilize a monitoring device which is installed in the vehicle by the driver. The monitoring device enables a monitoring feature in which it records audio and video data during rides in the vehicle. If necessary, such as for resolving a dispute or in the event of an accident, the monitoring device uploads the recorded data to a cloud storage backend. The monitoring device has a recognizable design and is integrated with mobility service provider applications used by both the driver of the vehicle and the riders who hail rides. The services of the monitoring device are operated by a trusted, neutral third-party. The trusted, neutral third-party may provide live qualified support at the touch of a button in an emergency or if the rider/driver simply feels uncomfortable or unsafe.

Abstract: An in-vehicle system for safe and supervised ride hailing experiences by a mobility service provider is disclosed. The system advantageously comprises a monitoring device which is installed in the vehicle by the driver. The monitoring device enables a monitoring feature in which it records audio and video data during rides in the vehicle. If necessary, such as for resolving a dispute or in the event of an accident, the monitoring device uploads the recorded data to a cloud storage backend. The monitoring device has a recognizable design and is integrated with mobility service provider applications used by both the driver of the vehicle and the riders who hail rides. The services of the monitoring device are operated by a trusted, neutral third-party. The trusted, neutral third-party may provide live qualified support at the touch of a button in an emergency or if the rider/driver simply feels uncomfortable or unsafe.

Abstract: A method and system are disclosed for monitoring passengers in within a cabin of a vehicle and determining whether the passengers are engaging in abnormal behavior. The method and system uses a novel vector to robustly and numerically represent the activity of the passengers in a respective frame, which is referred to herein as an “activity vector.” Additionally, a Gaussian Mixture Model is utilized by the method and system to distinguish between normal and abnormal passenger behavior. Cluster components of the Gaussian Mixture Model are advantageously learned using an unsupervised approach in which training data is not labeled or annotated to indicate normal and abnormal passenger behavior. In this way, the Gaussian Mixture Model can be trained at a very low cost.

Abstract: A sensing module for monitoring a cabin of a vehicle includes an environmental sensor configured to sense organic compounds in ambient air of the cabin and a particle sensor configured to detect particulate matter in the ambient air. The sensing module further includes a controller operably connected to the environmental sensor and the particle sensor and configured to receive sensor signals from the environmental sensor and the particle sensor and to transmit data to a remote server via the Internet. The sensing module has a housing configured to mount to a windshield of the vehicle. The housing supports the environmental sensor, the particle sensor, and the controller.

Abstract: A system and method are disclosed for monitoring rides in a vehicle in which a driver of the vehicle picks up a rider at a pickup location and drives the rider to a drop-off destination. The system includes at least one sensor arranged in the vehicle and configured to capture sensor data during the rides, a transceiver configured to communicate with a personal electronic device of a driver of the vehicle, a non-volatile memory configured to store data; and a processor. The system captures sensor data during a ride, determines an activity index indicating an activity level within the vehicle during the ride, and stores the sensor data captured during the ride with the activity index as metadata. In response to an upload request, the system uploads sensor data depending on the activity level.

Abstract: An in-vehicle sensing module for monitoring a vehicle is disclosed, which is advantageous for use in the context of a shared vehicle service, such as a car rental service, an autonomous taxi service, or a ride sharing service. The in-vehicle sensing module at least includes a controller, a cellular transceiver, and one or more integrated sensors configured to monitor a status of the vehicle. The in-vehicle sensing module utilizes appropriate algorithms, models, or thresholds to interpret sensor data and enrich the data with metadata and event detection. The in-vehicle sensing module uploads relevant sensor data, event data, or other metadata to a cloud storage backend, which is made accessible by authorized third-parties.

Abstract: A system for detecting and mitigating an unsafe condition in a vehicle includes an image sensor configured to generate and output image data of one or more seats in a cabin of the vehicle and a processing system operably connected to the image sensor and including at least one processor. The processing system is configured to receive the image data from the image sensor, process the image data to determine a location of at least one passenger in the cabin, detect that the at least one passenger is located outside of the one or more seats based on the determined location of the at least one passenger in the cabin, and operate at least one component of the vehicle in a predefined manner in response to detecting that the at least one passenger is located outside of the one or more seats.

Abstract: A system for detecting and resolving an abnormal settling event in a vehicle includes an image sensor and a processing system. The processing system is configured to, upon commencement of a passenger entry into the vehicle, receive image data from the image sensor, process the image data to determine a location of at least one passenger in the cabin, detect that the at least one passenger is located outside of the one or more seats of the vehicle, after a first predetermined time period has elapsed from the commencement of the passenger entry into the vehicle, based on the determined location of the at least one passenger in the cabin, and operate at least one component of the vehicle in a predefined manner in response to detecting that the at least one passenger is located outside of the one or more seats after the first predetermined time period has elapsed.

Abstract: A system and method are disclosed for monitoring rides in a vehicle in which a driver of the vehicle picks up a rider at a pickup location and drives the rider to a drop-off destination. The system includes at least one sensor arranged in the vehicle and configured to capture sensor data during the rides, a transceiver configured to communicate with a personal electronic device of a driver of the vehicle, a non-volatile memory configured to store data; and a processor. The system captures sensor data during a ride, receives a ride identifier from the personal electronic device that uniquely identifies the ride, and stores the sensor data captured during the ride with the ride identifier as metadata.

Abstract: A system and method are disclosed for monitoring rides in a vehicle in which a driver of the vehicle picks up a rider at a pickup location and drives the rider to a drop-off destination. The system includes at least one sensor arranged in the vehicle and configured to capture sensor data during the rides, a transceiver configured to communicate with a personal electronic device of a driver of the vehicle, a non-volatile memory configured to store data; and a processor. The system captures sensor data during a ride, determines an activity index indicating an activity level within the vehicle during the ride, and stores the sensor data captured during the ride with the activity index as metadata. In response to an upload request, the system uploads sensor data depending on the activity level.

Abstract: A method and system are disclosed for monitoring passengers in within a cabin of a vehicle and determining whether the passengers are engaging in abnormal behavior. The method and system uses a novel vector to robustly and numerically represent the activity of the passengers in a respective frame, which is referred to herein as an “activity vector.” Additionally, a Gaussian Mixture Model is utilized by the method and system to distinguish between normal and abnormal passenger behavior. Cluster components of the Gaussian Mixture Model are advantageously learned using an unsupervised approach in which training data is not labeled or annotated to indicate normal and abnormal passenger behavior. In this way, the Gaussian Mixture Model can be trained at a very low cost.

Abstract: A method and system are disclosed for monitoring passengers in within a cabin of a vehicle and determining whether the passengers are engaging in abnormal behavior. The method and system uses a novel vector to robustly and numerically represent the activity of the passengers in a respective frame, which is referred to herein as an “activity vector.” Additionally, a Gaussian Mixture Model is utilized by the method and system to distinguish between normal and abnormal passenger behavior. Cluster components of the Gaussian Mixture Model are advantageously learned using an unsupervised approach in which training data is not labeled or annotated to indicate normal and abnormal passenger behavior. In this way, the Gaussian Mixture Model can be trained at a very low cost.

Abstract: A cloud storage backend for managing data of in-vehicle sensing modules is disclosed, which is advantageous for use in the context of a shared vehicle service, such as a car rental service, an autonomous taxi service, or a ride sharing service. The in-vehicle sensing modules are configured to monitor a status of the vehicle and utilize appropriate algorithms, models, or thresholds to interpret sensor data and enrich the data with metadata and event detection. The cloud storage backend receives relevant sensor data, event data, or other metadata from the in-vehicle sensing modules and stores the data in a database which is made accessible by authorized third-parties.

Abstract: An in-vehicle sensing module for monitoring a vehicle is disclosed, which is advantageous for use in the context of a shared vehicle service, such as a car rental service, an autonomous taxi service, or a ride sharing service. The in-vehicle sensing module at least includes a controller, a cellular transceiver, and one or more integrated sensors configured to monitor a status of the vehicle. The in-vehicle sensing module utilizes appropriate algorithms, models, or thresholds to interpret sensor data and enrich the data with metadata and event detection. The in-vehicle sensing module uploads relevant sensor data, event data, or other metadata to a cloud storage backend, which is made accessible by authorized third-parties.

Abstract: A sensing module for monitoring a cabin of a vehicle includes an environmental sensor configured to sense organic compounds in ambient air of the cabin and a particle sensor configured to detect particulate matter in the ambient air. The sensing module further includes a controller operably connected to the environmental sensor and the particle sensor and configured to receive sensor signals from the environmental sensor and the particle sensor and to transmit data to a remote server via the Internet. The sensing module has a housing configured to mount to a windshield of the vehicle. The housing supports the environmental sensor, the particle sensor, and the controller.

Abstract: A method and system are disclosed for monitoring passengers in within a cabin of a vehicle and determining whether the passengers are engaging in abnormal behavior. The method and system uses a novel vector to robustly and numerically represent the activity of the passengers in a respective frame, which is referred to herein as an “activity vector.” Additionally, a Gaussian Mixture Model is utilized by the method and system to distinguish between normal and abnormal passenger behavior. Cluster components of the Gaussian Mixture Model are advantageously learned using an unsupervised approach in which training data is not labeled or annotated to indicate normal and abnormal passenger behavior. In this way, the Gaussian Mixture Model can be trained at a very low cost.

Abstract: A system for detecting and resolving an abnormal settling event in a vehicle includes an image sensor and a processing system. The processing system is configured to, upon commencement of a passenger entry into the vehicle, receive image data from the image sensor, process the image data to determine a location of at least one passenger in the cabin, detect that the at least one passenger is located outside of the one or more seats of the vehicle, after a first predetermined time period has elapsed from the commencement of the passenger entry into the vehicle, based on the determined location of the at least one passenger in the cabin, and operate at least one component of the vehicle in a predefined manner in response to detecting that the at least one passenger is located outside of the one or more seats after the first predetermined time period has elapsed.

Abstract: A system for detecting and mitigating an unsafe condition in a vehicle includes an image sensor configured to generate and output image data of one or more seats in a cabin of the vehicle and a processing system operably connected to the image sensor and including at least one processor. The processing system is configured to receive the image data from the image sensor, process the image data to determine a location of at least one passenger in the cabin, detect that the at least one passenger is located outside of the one or more seats based on the determined location of the at least one passenger in the cabin, and operate at least one component of the vehicle in a predefined manner in response to detecting that the at least one passenger is located outside of the one or more seats.