73results about "Control safety arrangements" patented technology

According to certain aspects, a computer-implemented method for operating an autonomous or semi-autonomous vehicle may be provided. With the customer's permission, an identity of a vehicle operator may be identified and a vehicle operator profile may be retrieved. Operating data regarding autonomous operation features operating the vehicle may be received from vehicle-mounted sensors. When a request to disable an autonomous feature is received, a risk level for the autonomous feature is determined and compared with a driver behavior setting for the autonomous feature stored in the vehicle operator profile. Based upon the risk level comparison, the autonomous vehicle retains control of vehicle or the autonomous feature is disengaged depending upon which is the safer driver- the autonomous vehicle or the vehicle human occupant. As a result, unsafe disengagement of self-driving functionality for autonomous vehicles may be alleviated. Insurance discounts may be provided for autonomous vehicles having this safety functionality.

This disclosure relates to apparatuses, systems, and methods for handling faults on vehicles. One or more processors in a vehicle may receive, from a control unit in response to a detection of a fault in the vehicle, an indication of a degradation in a performance constraint of the vehicle. The processors may determine, responsive to the degradation in the performance constraint, that the vehicle is unable to execute a set of commands to control a movement of the vehicle along a trajectory. The processors may generate, in accordance with the degradation in the performance constraint, a modified set of commands that the vehicle is able to execute to control the movement of the vehicle along at least a portion of one or more trajectories. The processors may provide the modified set of commands to the control unit of the vehicle to control the movement of the vehicle.

An evacuation running assistance system includes a road shoulder evacuation possibility determiner to determine if an own vehicle can be evacuated to a road shoulder; an own vehicle situation determiner to determine a current situation of an own vehicle in accordance with a time limit and the road shoulder evacuation possibility, a controller to control an own vehicle in accordance with the situation of the own vehicle; and a road shoulder evacuation possibility road determiner to acquire evacuation space information from a past running history of the own vehicle. The own vehicle situation determiner determines that the own vehicle is in the situation to be controlled to perform the on-lane stopping when the road shoulder evacuation possibility road determiner does not determine within the provisional time that the evacuation of the own vehicle to the road shoulder is possible.

In some embodiments, techniques are provided for analyzing time series data to detect anomalies. In some embodiments, a machine learning model is used to process time series data. In some embodiments, the machine learning model is trained in an unsupervised manner on a large amount of prior time series data, allowing a highly accurate model to be created from new data. In some embodiments, the training of the machine learning model alternates between a fit optimization and a prune optimization to allow a large amount of training data to be processed that includes unlabeled anomaly records. Because a machine learning model is used, anomalies within complex systems can be detected, including but not limited to autonomous vehicles such as unmanned aerial vehicles. When an anomaly is detected, a command can be sent to the monitored system (e.g., autonomous vehicle) to address the anomaly.

A vehicle control unit, including: a first control unit (2-1) with a microcomputer (15) that detects a state of an actuating switch (16) of an electrical component (3), monitors a signal indicating a state of an ignition key and controls the electrical component (3); and characterized by a second control unit (2-2) which is connected to the first control unit (2-1) via a vehicle-internal network (1a, 1b), where the microcomputer (15) when the actuating switch (16) of the electrical component (3) in the first control unit (2-1) is determined to be in an ON state for a longer period of time and the ignition key is determined to be in an OFF state, data transmission from the first control unit (2-1) to the second control unit (2-2) is interrupted, or the first control unit (2-1) is put into a receive-only mode, a driver of a vehicle is notified of a fault and performs a process to detect the cancellation of a longer-term on-state of the actuating switch (16), if the actuating switch (16) of the electrical component (3) is determined to be in the on-state for a longer period of time and the ignition key is determined to be in the on-state, and stores an error report in a memory device and interrupts the notification of the driver if the long-term on state of the actuating switch (16) persists continuously for a certain number of times or continuously for a certain period of time when the ignition key is in the off state after the process for detecting the cancellation of the long-term on state of the actuating switch (16).

An aerial vehicle includes a battery, a propulsion system including a motor and a rotor, and a flight controller configured to: in response to one or more parameters of the battery not satisfying a preset condition, cause the aerial vehicle to enter a warm up mode, in which the motor drives the rotor to rotate but the aerial vehicle is prevented from taking off, and while the aerial vehicle is in the warm up mode, in response to the one or more parameters of the battery satisfying the preset condition, allow the aerial vehicle to take off.

Systems and methods are provided for navigating an autonomous vehicle using reinforcement learning techniques. 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 navigational state associated with the host vehicle; provide the navigational state to a trained navigational system; receive, from the trained navigational system, a desired navigational action for execution by the host vehicle in response to the identified navigational state; analyze the desired navigational action relative to one or more predefined navigational constraints; determine an actual navigational action for the host vehicle, wherein the actual navigational action includes at least one modification of the desired navigational action determined based on the one or more predefined navigational constraints; and cause at least one adjustment of a navigational actuator of the host vehicle in response to the determined actual navigational action for the host vehicle.

A moving body control method controls a moving body having a function of operating in accordance with a remote instruction in a predetermined area. The moving body control method includes: a remote instruction verification process that determines whether or not the remote instruction received by the moving body is valid; and an operation limiting process that limits at least a part of an operation of the moving body without following the remote instruction, when the remote instruction received by the moving body is invalid. The remote instruction verification process includes: determining whether or not the moving body is present in the predetermined area at a time when the moving body receives the remote instruction; and determining that the remote instruction is invalid, when the moving body is not present in the predetermined area at the time when the moving body receives the remote instruction.

Systems and techniques for reliable real-time deployment of robot safety updates are described herein. Condition data may be collected for an environment in which a robot is operating. The robot may be classified with a condition type. Condition type data selected from the condition data may be analyzed based on the condition type to calculate a safety risk level for the robot. A microservice may be identified to provide a robot safety rule for the robot based on the safety risk level. The microservice may be identified using a safety prediction model generated from the condition type data.

Methods and systems for monitoring use, determining risk, and pricing insurance policies for a vehicle having one or more autonomous or semi-autonomous operation features are provided. According to certain aspects, the operating status of the features, the identity of a vehicle operator, risk levels for operation of the vehicle by the vehicle operator, or damage to the vehicle may be determined based upon sensor or other data. According to further aspects, decisions regarding transferring control between the features and the vehicle operator may be made based upon sensor data and information regarding the vehicle operator. Additional aspects may recommend or install updates to the autonomous operation features based upon determined risk levels. Some aspects may include monitoring transportation infrastructure and communicating information about the infrastructure to vehicles.

The application provides a multi-sensor synchronous control method and device, an unmanned system and a medium, comprising: determining a delay duration of a to-be-synchronized sensor based on coordinates of a preset reference point corresponding to the to-be-synchronized sensor and control characteristics of the to-be-synchronized sensor; determining a trigger time of the to-be-synchronized sensor based on a preset target sampling time and the delay duration; and triggering a synchronization clock signal corresponding to the to-be-synchronized sensor by using the trigger time, so that the to-be-synchronized sensor samples data of the preset reference point at the target sampling time. In this way, different trigger times can be set for each sensor, and all sensors can sample corresponding reference points at the preset target sampling time, realizing synchronous sampling of all sensors on the same object, improving the accuracy of the sensing result of the sensing system, and further improving the stability and safety of the unmanned system.

Method (200) for setting a control signal (135) for controlling at least one safety device (130) for a vehicle (100), wherein the method (200) comprises the following steps: Reading (210) a driving mode signal (115) which has a first value when a manual driving mode of the vehicle (100) is set, and which has a second value when at least a partially automatic driving mode of the vehicle (100) is set;and adjusting (220) at least one parameter of the control signal (135) for controlling the at least one safety device (130) using the driving mode signal (115) to set the control signal (135), characterized in that, when responding to the first value of the driving mode signal (115) read in the reading step (210), the at least one parameter is set to a first parameter value in the adjusting step (220), wherein the first parameter value causes a partial triggering of the at least one safety device (130) when the at least one safety device (130) is controlled.

An autonomous vehicle safety system, a method of operating the same and an autonomous vehicle equipped with the autonomous vehicle safety system. An autonomous vehicle Bluetooth receiver is connected in a signal transmitting fashion to the engine control unit of an autonomous vehicle. Upon receiving a signal from a remote Bluetooth transmitter at a predefined proximity distance between the transmitter and the receiver, a microcontroller connected to the receiver actuates a relay switch disabling the engine control unit. The engine control unit may then also engage the vehicle breaks, lock the steering in place, and only after not receiving a proximal Bluetooth signal for a particular time span, enable again the autonomous vehicle operating mode.

Systems and methods are provided for navigating an autonomous vehicle using reinforcement learning techniques. 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 navigational state associated with the host vehicle; provide the navigational state to a trained navigational system; receive, from the trained navigational system, a desired navigational action for execution by the host vehicle in response to the identified navigational state; analyze the desired navigational action relative to one or more predefined navigational constraints; determine an actual navigational action for the host vehicle, wherein the actual navigational action includes at least one modification of the desired navigational action determined based on the one or more predefined navigational constraints; and cause at least one adjustment of a navigational actuator of the host vehicle in response to the determined actual navigational action for the host vehicle.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for an unmanned aerial system inspection system. One of the methods is performed by a UAV and includes obtaining, from a user device, flight operation information describing an inspection of a vertical structure to be performed, the flight operation information including locations of one or more safe locations for vertical inspection. A location of the UAV is determined to correspond to a first safe location for vertical inspection. A first inspection of the structure is performed is performed at the first safe location, the first inspection including activating cameras. A second safe location is traveled to, and a second inspection of the structure is performed. Information associated with the inspection is provided to the user device.

The invention provides a multi-sensor synchronization control method and device, an autonomous driving system, and a medium. The method includes: determining delay duration of sensors to be synchronized based on coordinates of a preset reference point corresponding to the sensors to be synchronized and control characteristics of the sensors to be synchronized; determining a triggering moment of the sensors to be synchronized based on a preset target sampling moment and the delay duration; and triggering, by using the triggering moment, a synchronous clock signal corresponding to the sensors to be synchronized, so that the sensors to be synchronized sample data on the preset reference point at the target sampling moment. In this way, a different triggering moment can be set for each sensor, and all the sensors can perform sampling on corresponding reference points at the preset target sampling moment, thereby implementing synchronous sampling on the same object by all the sensors, improving the accuracy of a perception result of a perception system, and thus improving the stability and safety of the autonomous driving system.

Achieved is a configuration in which a reliability of self-position estimation of a mobile device, such as a drone, is calculated to perform warning display and flight control. A sensor data acquisition unit that acquires detection information of a sensor provided in a mobile body, a self-position estimation unit that receives an input of the detection information of the sensor from the sensor data acquisition unit and executes self-position estimation of the mobile body, an environment information analysis unit that analyzes environment information such as ambient brightness when a self-position estimation process has been executed in the self-position estimation unit and a score of a feature point of a camera-captured image, and an estimated self-position reliability calculation unit that calculates a reliability of a self-position estimated by the self-position estimation unit on the basis of the environment information analyzed by the environment information analysis unit are provided to perform warning display and flight control on the basis of the calculated reliability.

A wireless control system comprises a plurality of local stations linked by a communication network. Each local station transmits, in a respective radio coverage area, an enduring status signal. An autonomous vehicle is authorized to move while it receives the status signal. When an emergency stop switch of the local station is activated, the local station interrupts its transmission of the status signal. It also instructs one or more further local stations, to which it is linked by a communication network, to interrupt their transmission of the status signal. In this way, the activation of the local emergency stop switch will have effect throughout the control system and will eventually bring all autonomous vehicles to a halt.

Herein, a "perception statistical performance model" (PSPM) for modelling a perception slice of a runtime stack for an autonomous vehicle or other robotic system may be used e.g. for safety / performance testing. A PSPM is configured to: receive a computed perception ground truth t; determine from the perception ground truth t, based on a set of learned parameters, a probabilistic perception uncertainty distribution of the form p(e|t), p(e|t,c), in which p(e|t,c) denotes the probability of the perception slice computing a particular perception output e given the computed perception ground truth t and the one or more confounders c, and the probabilistic perception uncertainty distribution is defined over a range of possible perception outputs, the parameters learned from a set of actual perception outputs generated using the perception slice to be modelled, wherein each confounder is a variable of the PSPM whose value characterized a physical condition on which p(e|t,c) depends.

Herein, a "perception statistical performance model" (PSPM) for modelling a perception slice of a runtime stack for an autonomous vehicle or other robotic system may be used e.g. for safety / performance testing. A first PSPM is configured to: receive a computed perception ground truth; determine from the perception ground truth, based on a set of learned parameters, a probabilistic perception uncertainty distribution, the parameters learned from a set of actual perception outputs generated using the perception slice to be modelled, in order to compute a first time series of perception outputs. A second time series of perception outputs is computed using a second PSPM for modelling a second perception slice of the runtime stack, the first PSPM learned from data of a first sensor modality of the perception slice and the time series, and the second PSPM learned independently thereof from data of a second sensor modality of the second perception slice and the second time series.

A lifting and / or transporting arrangement includes a remote-controllable lifting and / or transporting device, such as an aircraft pusher; a remote-control transmitter for controlling a lifting and / or transporting movement of the lifting and / or transporting device; and a safety apparatus on which data for defining a horizontally extending virtual enablement area are stored and which has a sensor system for determining a position of the remote-control sensor. A control process of the remote-control transmitter can be enabled or blocked by the safety apparatus on the basis of a currently detected position of the remote-control transmitter with respect to the enablement area. The enablement area has a limited vertical extent in a vertical direction, and the sensor system of the safety apparatus has height determination device for determining a height of the remote-control transmitter. The control process can be enabled or blocked on the basis of a currently detected height of the remote-control transmitter with respect to the vertical extent of the enablement area.

System (100) for automated driving, comprising: an obstacle detection unit (11) configured to detect at least one obstacle around a vehicle based on the vehicle's surroundings; a timetable generation unit (12) which is configured to generate a timetable for the vehicle based on the environment and a result of the detection of at least one obstacle; a vehicle control device (13) which is configured to control the vehicle based on the timetable; an overwrite unit (14) configured to perform an overwrite operation in response to an input from a user of the vehicle in order to modify the result of the detection of at least one obstacle, which is used to generate the timetable; and a presentation unit (21) which is configured to present to the user the result of the detection of at least one obstacle, wherein the overwrite unit (14) is configured such that it performs the overwrite operation only during a period of time in which input is continued by the user, wherein the overwrite unit (14) is configured such that it performs an operation as the overwrite operation to exclude at least part of the at least one obstacle from the result of the detection of the at least one obstacle, wherein the presentation unit (21) is configured to serve as an input unit which receives input from the user, wherein the overwrite unit (14) is configured to perform the overwrite operation according to the input from the user at the input unit, and wherein the input unit comprises a touch sensor or a touch panel.

An impairment analysis (“IA”) computer system for alerting a first driver of a first vehicle to a driving hazard posed by a second vehicle operated by a second driver is provided. The IA computer system is associated with the first vehicle, and includes at least one processor in communication with at least one memory device. The at least one processor is programmed to: (i) receive second vehicle data including second driver data and second vehicle condition data, where the second vehicle data is collected by a plurality of sensors included on the first vehicle; (ii) analyze the second vehicle data by applying a baseline model to the second vehicle data; (iii) determine that the second vehicle poses a driving hazard to the first vehicle based upon the analysis; and / or (iv) generate an alert signal based upon the determination that the second vehicle poses a driving hazard to the first vehicle.

The present teachings provide a non-transitory computer-readable storage medium, including a processor-executable routines that, when executed by a processor, facilitate a performance of operations. The operations include detecting a condition using one or more sensors of a vehicle, wherein the vehicle includes motors. The operations include signaling the condition by causing one of the motors of the vehicle to produce an audible tone. The operations include defining an operating margin for the one of the motors to produce the audible tone. The operations include changing an amount of operating current supplied to the one of the motors based on a maximum amount of current that can be supplied to the one of the motors based on the operating margin.

In various examples, incremental booting of functions for autonomous systems and applications is described herein. Systems and methods are disclosed that boot functions of a machine using different phases in order to more quickly provide at least a portion of the functions without compromising the safety of the machine. For example, during a first phase, resources associated with the machine may be allocated while the functions are initialized. Additionally, at least a first portion of the functions, such as functions that do not compromise the safety of the machine if errors occur, may be made available for use by one or more users of the machine. Next, during a second phase, at least a second portion of the functions, such as functions that may compromise the safety of the machine if errors occur, may be made available for use by the user(s) of the machine.

A control device (100) is provided with: a communication unit (110) that receives an external start signal transmitted from the outside of an autonomous vehicle (MV) for starting autonomous driving of the autonomous vehicle (MV); a control unit (140) that performs a process necessary for automatic driving; and a person determination unit (120) that determines whether or not a person is present in the cabin of the autonomous vehicle. When the person determination unit determines that a person is present in the vehicle interior, the control unit does not start automatic driving even if the communication unit receives the external start signal.

To provide a technique for enabling a remote operator to operate a vehicle in the same way as an operation of a driver in the vehicle.SOLUTION: A vehicle control system for controlling a vehicle based on an operation amount to be input by a remote operator includes one or more processors. In a case where a stop holding function to hold a stop state of the vehicle is actuated, the one or more processors continue an actuation of the stop holding function when an input of an acceleration operation is not received from the remote operator, and cancel the actuation of the stop holding function when an input of an acceleration operation is received from the remote operator.SELECTED DRAWING: Figure 6

A "perception statistical performance model" (PSPM) can be used herein to model a perception slice of a runtime stack of an autonomous vehicle or other robotic system, e.g., for safety / performance testing. The PSPM is configured to: receive a computed perception ground truth t; determine, from the perception ground truth t, a probabilistic perception uncertainty distribution of the form p(e|t), p(e|t,c), based on a set of learned parameters, where p(e|t,c) represents a probability of the perception slice computing a particular perception output e given the computed perception ground truth t and one or more confounders c, and the probabilistic perception uncertainty distribution is defined over a range of possible perception outputs, the parameters learned from a set of actual perception outputs generated using the perception slice to be modeled, where each confounder is a variable of the PSPM, the value of the variable characterizing a physical condition, p(e|t,c) dependent on the variable.

Herein, a "perception statistical performance model" (PSPM) for modelling a perception slice of a runtime stack for an autonomous vehicle or other robotic system may be used e.g. for safety / performance testing. A PSPM is configured to: receive a computed perception ground truth; determine from the perception ground truth, based on a set of learned parameters, a probabilistic perception uncertainty distribution, the parameters learned from a set of actual perception outputs generated using the perception slice to be modelled. The modelled perception slice includes an online error estimator, and the computer system is configured to use the PSPM to obtain a predicted online error estimate for the perception output in response to the perception ground truth. This recognizes that online perception error estimates may, themselves, be subject to error.