A human-robot collaboration system, including at least one processor; and a non-transitory computer-readable storage medium including instructions that, when executed by the at least one processor, cause the at least one processor to: predict a human atomic action based on a probability density function of possible human atomic actions for performing a predefined task; and plan a motion of the robot based on the predicted human atomic action.

Methods and systems for assessing, detecting, and responding to malfunctions involving components of autonomous vehicles and/or smart homes are described herein. Autonomous operation features and related components can be assessed using direct or indirect data regarding operation. Vehicle collision and/or smart home incident monitoring, damage detection, and responses are also described, with particular focus on the particular challenges associated with incident response for unoccupied vehicles and/or smart homes. Operating data associated with the autonomous vehicle and/or smart home may be received. Within the operating data, a divergence between sensor data from one or more sensors and control data from one or more autonomous operation features may be identified. Based on the divergence, it may be determined that an incident has occurred. Accordingly, a response to the incident may be determined. The response may be implemented by the autonomous vehicle and/or smart home.

A system uses a machine learning based model to determine attributes describing states of mind and behavior of traffic entities in video frames captured by an autonomous vehicle. The system classifies video frames according to traffic scenarios depicted, where each scenario is associated with a filter based on vehicle attributes, traffic attributes, and road attributes. The system identifies a set of video frames associated with ground truth scenarios for validating the accuracy of the machine learning based model and predicts attributes of traffic entities in the video frames. The system analyzes video frames captured after the set of video frames to determine actual attributes of the traffic entities. Based on a comparison of the predicted attributes and actual attributes, the system determines a likelihood of the machine learning based model making accurate predictions and uses the likelihood to generate a navigation action table for controlling the autonomous vehicle.

A method for predicting at least one future velocity vector and/or a future pose of a pedestrian in an area of prediction. A map of a surrounding environment of the pedestrian and current velocity vectors of other pedestrians in the area of prediction are taken into account in the prediction.

Methods and systems for assessing, detecting, and responding to malfunctions involving components of autonomous vehicles and/or smart homes are described herein. Malfunctions may be detected by receiving sensor data from a plurality of sensors. One of these sensors may be selected for assessment. An electronic device may obtain from the selected sensor a set of signals. When the set of signals includes signals that are outside of a determined range of signals associated with proper functioning for the selected sensor, it may be determined that the selected sensor is malfunctioning. In response, an action may be performed to resolve the malfunction and/or mitigate consequences of the malfunction.

One or more techniques and/or systems are provided for notifying drivers to assume manual vehicle control of vehicles. For example, sensor data is acquired from on-board vehicles sensors (e.g., radar, sonar, and/or camera imagery of a crosswalk) of a vehicle that is in an autonomous driving mode. In an example, the sensor data is augmented with driving condition data aggregated from vehicle sensor data of other vehicles (e.g., a cloud service collects and aggregates vehicle sensor data from vehicles within the crosswalk to identify and provide the driving condition data to the vehicle). The sensor data (e.g., augmented sensor data) is evaluated to identify a driving condition of a road segment, such as the crosswalk (e.g., pedestrians protesting within the crosswalk). Responsive to the driving condition exceeding a complexity threshold for autonomous driving decision making functionality, a driver alert to assume manual vehicle control may be provided to a driver.

The present disclosure relates to systems and methods for host vehicle navigation. In one implementation, a navigation system for a host vehicle may include at least one processing device programmed to receive, from a camera, a plurality of images representative of an environment of the host vehicle; analyze the plurality of images to identify a first flow of traffic and a second flow of traffic; determine a presence of at least one navigational state characteristic indicative of an alternating merging of the first flow of traffic and the second flow of traffic into a merged lane; cause at least a first navigational change to allow one target vehicle from the first flow of traffic to proceed ahead of the host vehicle; and cause at least a second navigational change to cause the host vehicle to follow the target vehicle into the merged lane.

A method for automated parking of a vehicle includes providing image data captured by at least one vision sensor of the vehicle to an electronic control unit (ECU), and providing a parking scene map to the ECU. A free parking space present in the parking scene map is selected as a target parking space, and a parking path from a current vehicle location to the target parking space is formulated. The vehicle is autonomously maneuvered along the parking path towards the target parking space. Responsive to detection of a pedestrian in the target parking space or along the parking path, the vehicle is stopped until the detected pedestrian moves out of the target parking space or out of the parking path. After the detected pedestrian has moved out of the target parking space or out of the parking path, the vehicle continues autonomous maneuvering along the parking path towards the target parking space.

Aspects of the disclosure relate to generating a speed plan for an autonomous vehicle. As an example, the vehicle is maneuvered in an autonomous driving mode along a route using pre-stored map information. This information identifies a plurality of keep clear regions where the vehicle should not stop but can drive through in the autonomous driving mode. Each keep clear region of the plurality of keep clear regions is associated with a priority value. A subset of the plurality of keep clear regions is identified based on the route. A speed plan for stopping the vehicle is generated based on the priority values associated with the keep clear regions of the subset. The speed plan identifies a location for stopping the vehicle. The speed plan is used to stop the vehicle in the location.

Methods and systems for autonomous and semi-autonomous vehicle routing are disclosed. Roadway suitability for autonomous operation is scored to facilitate use in route determination. Maps of roadways suitable for various levels of autonomous operation may be generated. Such map data may be used by autonomous vehicles or other computer devices in determining routes based upon criteria for vehicle trips. Such routes may be automatically updated based upon changes in road conditions, vehicle conditions, operator conditions, or environmental conditions. Emergency routing using such map data is described, such as automatic routing and travel when a passenger is experiencing a medical emergency.