301 results about "Gps data" patented technology

The invention discloses an unmanned aerial vehicle trajectory tracking system based on multi-sensor fusion, which relates to the technical field of unmanned system autonomous navigation and control, and comprises an unmanned aerial vehicle data acquisition module, an unmanned aerial vehicle feature extraction module, a trajectory prediction deviation scoring module, a trajectory dynamic correction module and a trajectory monitoring module. The unmanned aerial vehicle data acquisition module acquires and preprocesses unmanned aerial vehicle and GPS data, the unmanned aerial vehicle feature extraction module constructs a trajectory tracking feature sequence table, the trajectory prediction deviation scoring module outputs three-dimensional trajectory prediction coordinates and deviation scores, and the trajectory dynamic correction module generates correction parameters. Based on the judgment logic, the judgment logic is converted into an unmanned aerial vehicle correction instruction and attitude feedback quantity, and the track monitoring module constructs a visual interface. According to the method, the trajectory tracking precision of the unmanned aerial vehicle in a complex environment is remarkably improved, real-time detection and dynamic correction of abnormal deviation are realized, the state monitoring capability is enhanced by a three-dimensional visual interface, the trajectory tracking error is reduced, the misjudgment rate is reduced, and the flight safety and the control efficiency are guaranteed.

The invention relates to the technical field of aviation communication, and discloses a true north calibration system and method for an aviation airborne laser communication equipment alignment system, and the method comprises the steps: synchronously collecting flight data through an IMU sensor and a double-antenna GPS at a preset sampling frequency, and outputting the real-time attitude angle information of an aircraft through an IMU; filtering and de-noising the synchronously acquired IMU and double-antenna GPS data, and carrying out noise suppression and deviation compensation on accelerometer and gyroscope data output by the IMU by adopting a Kalman filtering algorithm; based on the preprocessed sensor data, calculating an initial course angle by measuring relative position change and phase difference between antennas by using a double-antenna GPS (Global Positioning System); the calibration controller transmits the calculated true north direction information to an alignment system of the laser communication equipment in real time through a communication interface; and the attitude change and external environment disturbance of the aircraft are continuously monitored in the whole flight process. The method has the advantage that the alignment performance and the communication stability of the laser communication equipment are improved.

The invention relates to the technical field of image processing, in particular to an irrigation prescription map inversion method based on unmanned aerial vehicle spectrum data. The method comprises the following steps: carrying out unmanned aerial vehicle data acquisition on a farmland area to obtain a multispectral image, DEM data and meteorological interpolation data; performing micro-landform parameter calculation according to the multispectral image, DEM data and unmanned aerial vehicle GPS data, and constructing a terrain-spectrum coupling model to obtain a terrain relief degree map and the terrain-spectrum coupling model; performing vegetation multi-dimensional state analysis by using a terrain-spectrum coupling model to obtain a vegetation index map, a canopy temperature map and a water stress preliminary judgment map; performing net radiation calculation by using the meteorological interpolation data to obtain a net radiation diagram; according to the net radiation diagram, the vegetation index diagram and the topographic relief degree diagram, soil heat flux estimation is conducted, and a soil heat flux diagram is obtained. According to the method, collaborative inversion is carried out by eliminating the terrain effect, and the precision and reliability of irrigation prescription map inversion are improved.

The invention discloses a shared bicycle dangerous behavior identification method based on GPS trajectory data tracking, and belongs to the technical field of traffic safety. Comprising the steps of abnormal point elimination, point supplementation, coordinate system unification and normalization so as to improve data accuracy, and clustering division is carried out on data according to data features; in combination with a road network and POI data, hotspot and non-hotspot areas are determined through track intensity analysis; six dangerous behaviors are defined, historical data are analyzed, a threshold value is set, a predefined method is used for recognizing dangerous driving behaviors, and finally a recognition model is established to achieve accurate recognition of the six dangerous behaviors. Through mathematical statistics, the riding characteristics and dangerous behavior characteristics of the shared bicycle under different conditions are compared and researched; according to the dangerous behavior identification method based on the shared bicycle GPS track data, the utilization rate of the shared bicycle GPS data can be effectively improved, dangerous behaviors can be accurately identified, and the safety of riders and other traffic participants is guaranteed.

The invention relates to the technical field of geo-fence triggering, in particular to a multi-mode positioning data fused GEO fence triggering method and triggering system. The method comprises the following steps that according to the distance between equipment and the fence boundary, a layered positioning strategy is adopted to demarcate a safe area, a transition area and a critical area; when the device is located in the safe area, outputting a coarse position coordinate of the device and a corresponding coarse positioning confidence coefficient by adopting a low-power-consumption coarse positioning algorithm; when the equipment is located in the transition area, fusing the GPS data and the IMU data by adopting an adaptive fusion algorithm to obtain an intermediate fusion position, an intermediate fusion speed and an intermediate positioning confidence coefficient; fusing the multi-modal data when the equipment is located in the critical zone, and predicting the probability that the equipment invades the fence by combining a trajectory risk assessment algorithm based on the equipment acceleration and the distance between the equipment and the fence boundary. Through hierarchical positioning strategy and multi-modal data fusion, the problems of insufficient precision and high power consumption of a single positioning mode in a complex scene are solved.

A system and method attest the veracity of audiovisual depictions by inferring motion of the device recording those depictions from the audiovisual data, and comparing this inferred motion with contemporaneously sampled sensor data from that device, such as accelerometer or GPS data. If the inferred motion sensibly corresponds to changes in the other sensor data, then the video is determined to be authentic. Otherwise, an anomalous condition is signaled, altering the viewer that the video may have been digitally altered.

The invention relates to the field of rocket recovery pose estimation design, and particularly discloses a multi-mode cooperative perception driven rocket recovery high-precision pose estimation system, which comprises a multi-scale Marker layout module, a multi-scale cooperative perception driven rocket recovery high-precision pose estimation module and a multi-scale cooperative perception driven rocket recovery high-precision pose estimation module, and is characterized in that the multi-scale Marker layout module is used for arranging three-level height detection bands around a recoverer device, and each band is configured with Marker type layout modes of different sizes; the multi-view cooperative detection module is used for collecting images through a plurality of high-resolution cameras installed on the side face of the rocket and outputting visual poses in combination with scale adaptive detection process optimization and a space consistency constraint strategy; the multi-sensor fusion module is used for performing fusion optimization on the visual pose, the IMU and the GPS data through a factor graph to construct a factor graph joint optimization framework; accuracy and stability of a rocket recovery pose estimation system are realized, and real-time performance, accuracy and anti-interference capability of a rocket recovery pose estimation scheme are improved.

In some implementations, a controller may obtain global positioning system (GPS) data from a GPS unit of an enrichment device. The controller may determine habitat information identifying a habitat for animals, of a plurality of habitats of animals, using the GPS data. The controller may determine a geofence of the habitat for animals based on the habitat information. The controller may determine a movement profile associated with the habitat for animals using the habitat information. The movement profile identifies one or more motions of the enrichment device. The controller may cause the enrichment device to move, within a boundary of the geofence, in accordance with the one or more motions of the movement profile. The enrichment device may move in accordance with the one or more motions to stimulate movement of one or more animals of the habitat for animals. The controller may monitor the movement and / or animal interaction.

The invention is suitable for the technical field of electric vehicle energy consumption management, and provides an electric vehicle energy consumption prediction method and system using GPS data, and the method comprises the steps: obtaining a GPS original data stream in real time, carrying out the preprocessing, and fusing a digital elevation model to generate an altitude sequence; calculating an instantaneous slope value and a slope change rate based on the altitude sequence, and quantifying the topographic relief intensity; calculating additional acceleration and additional energy consumption by combining the gradient change rate; based on the altitude sequence, the speed data and the real-time state of the kinetic energy recovery system, recoverable electric energy is predicted; and combining the reference energy consumption, the additional energy consumption and the recyclable electric energy to calculate and predict net energy consumption, and triggering model dynamic calibration through a continuous error. According to the invention, the energy consumption prediction precision is improved, and reliable support is provided for electric vehicle energy management.

The invention discloses a traffic jam detection and early warning method based on an unmanned aerial vehicle video stream, and belongs to the technical field of intelligent traffic. The system is combined with a floating vehicle triggering model, collects road videos in real time through an unmanned aerial vehicle, and realizes high-precision vehicle detection and tracking in combination with improved YOLOv8 and DeepSort algorithms; unmanned aerial vehicle data and floating vehicle GPS data are fused, three elements of traffic flow are quantified, and a dual-mechanism congestion detection model is constructed; respectively utilizing a GRU cooperative traffic wave theory to predict the congestion dissipation duration and utilizing an improved GRU + GCN model to predict the traffic situation; the signal lamps are dynamically regulated and controlled through multi-agent reinforcement learning, and active congestion relieving is achieved. The method has the advantages of wide-area coverage, high-precision perception and intelligent response, and the urban traffic control efficiency is remarkably improved.

The invention relates to a manual driving vehicle overspeed behavior identification and intervention method based on a detection carrier, and the method comprises the steps: building a detection carrier system, collecting adjacent HV point cloud data through a laser radar, carrying out the median average filtering, K-Means and DBSCAN combined clustering preprocessing, and achieving the multi-target dynamic tracking through a Point Pill + DeepSORT model; and calculating the real average speed of the HV by using a 1-3s self-adaptive speed measurement window, and comparing the road speed limit to complete the overspeed identification. And when overspeed is judged, dynamically triggering a corresponding high-speed camera to capture, extracting license plate information, fusing speed and GPS data to generate illegal evidence. The method comprises the following steps: quantifying the influence of a detection carrier on an HV individual overspeed behavior, training an HV swarm agent based on VAE-WGAN data enhancement and a double-branch deep Q network, simulating multiple traffic scenes in SUMO, and determining an optimal deployment scheme of the detection carrier by using a dynamic Gaussian mixture Bayesian network and a double-layer double-target optimization model. The method breaks through the limitation of traditional speed measurement, improves the recognition precision and deterrence effect, and balances the traffic safety and efficiency.

The invention belongs to the technical field of intelligent hidden danger detection, and more specifically relates to a hidden danger target geographic coordinate accurate calculation method and system based on an unmanned aerial vehicle and artificial intelligence detection, and a computer readable storage medium. The method comprises the following steps: the unmanned aerial vehicle identifies a target in a video image in real time and analyzes a video frame; when analyzing that the current frame image contains the hidden danger target, the control interface notifies the unmanned aerial vehicle to hover; according to the pixel coordinates of the hidden danger target in the video image and the camera focal length, the camera negative film width and height and the imaging size of the unmanned aerial vehicle cradle head, calculating a cradle head relative adjustment angle which enables the hidden danger target to be centered; according to the obtained relative adjustment angle of the holder, adjusting the holder to enable the hidden danger target to be in the center of a holder picture; and after the adjustment is completed, calculating the GPS data of the hidden danger target. The problems that in the prior art, data processing lags behind, hidden dangers cannot achieve closed-loop management, system fragmentation lacks end-to-end collaborative optimization, and consequently the overall efficiency is limited are solved.

The invention discloses a traffic flow prediction method, system and device, a medium and a computer program product, and the method comprises the steps: inputting a training set into a traffic flow prediction model, training the model, and obtaining a trained traffic flow prediction model; the trained traffic flow prediction model comprises a pre-trained spatial-temporal feature information extraction network used for performing feature extraction on historical GPS data to obtain spatial-temporal features of a known region; the multi-source information fusion flow prediction module is used for carrying out feature extraction on the historical traffic flow data to obtain traffic flow features of a known region, and is also used for carrying out feature extraction on the meteorological data to obtain meteorological data features of the known region; and the fusion prediction module is used for performing feature fusion on the spatial-temporal features, the traffic flow features and the meteorological data features of the known region, and predicting the fused features to obtain a traffic flow prediction value of the known region.

The invention, which relates to the technical field of mechanism material distribution, discloses a mechanism material distribution real-time collaboration and traceability system based on a unified data platform, comprising a data specification module, a dynamic routing module, a traceability code generation module, a storage verification module, a nutrition calculation module, a report generation module and an abnormal response module. The data specification module adopts a unified JSON Schema to define data specifications of a supplier end production batch number and a quality inspection report, a distribution end warehouse temperature and humidity and a vehicle GPS, and a school end acceptance image and student feedback; the dynamic routing module is used for realizing state change and pushing to the three-party system in real time through a dynamic routing engine based on the output of the data specification module; the traceability code generation module is used for generating a four-segment traceability code according to the structured data of the data specification module and storing a key data hash value of each link; and the evidence storage verification module is based on the output of the traceability code generation module.

A method and system for adjusting a digital elevation model (DEM). A satellite image with one or more roads and the DEM are received. The DEM includes an elevation of each pixel of the satellite image. Recorded GPS data corresponding to the one or more roads is received. A pixel in the satellite image corresponding to the one or more roads is determined. A determination is made as to whether the location of the pixel corresponds to the GPS data. After determining that the location of the pixel does not correspond to the GPS data, a set of pixels is determined in the DEM that surround the pixel. The elevation of each pixel of the set of pixels of the DEM is adjusted so that the location of the pixel in the satellite image corresponds to the one or more roads.

A method for collecting road geometry information includes obtaining gyroscope data from a gyroscope of a personal device fixed to a moving vehicle; aligning the gyroscope data with GPS data from a GPS module of the personal device to generate aligned data; matching the aligned data to a nearest road segment based on a preexisting map; determining a time of stable dynamics of the vehicle based on the map and the matched nearest road segment; and correcting a drift of a gyroscope using data from an accelerometer of the personal device during the determined time of stable dynamics.

The invention provides an intelligent driving scene reconstruction method and system and a storage medium, and the method comprises the steps: obtaining an OpenStretMap map file based on the real vehicle GPS data, and carrying out the lane-level road network matching based on the OpenStretMap map file; the OpenStreetMap map file obtained after matching of the lane level road network is converted into an OpenDR IVE map file; based on real vehicle GPS data, generating a main vehicle track by using a FastDTW algorithm; based on the environment vehicle information, generating a dynamic interaction object track through a Hungary matching algorithm; and fusing the OpenDRIVE map file, the main vehicle track and the dynamic interaction object track, and reconstructing a scene file in an OpenSCENARIO format in preset simulation software. According to the method, automatic reconstruction is carried out by utilizing the real vehicle data, the real vehicle data can be automatically stored as a universal OpenSCENARIO scene file, simulation testing can be carried out in different types of automatic driving simulation software, a real driving scene can be accurately simulated, and a reliable simulation environment is provided for development, testing and verification of an intelligent driving algorithm.

The invention relates to a roadside sensor calibration method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a plurality of pieces of global positioning system (GPS) data and a plurality of pieces of sensor data in the driving process of a target vehicle, the GPS data being collected by a GPS system of the target vehicle, and the sensor data being obtained by tracking the target vehicle through a roadside sensor; time alignment processing is carried out on the GPS data and the sensor data, multiple groups of point pairs are determined, and each group of point pairs comprises the GPS data and the sensor data corresponding to the same time point; based on the mapping relation between the GPS data and the sensor data in each group of point pairs, first external parameters between the GPS system and the sensors are determined, and the first external parameters are used for calibrating the sensors. According to the method, calibration efficiency can be improved, regular and even dynamic calibration updating can be conveniently achieved, actual requirements in a complex roadside environment are met, and the method can be suitable for scenes where sensors are far in distance and different in orientation in an intelligent traffic system.

The invention relates to the technical field of underway systems, in particular to a mobile underway analysis visualization system. According to the technical scheme, the system comprises a data acquisition layer, a data processing and analysis layer, a software integration layer, a function application layer and a user interaction layer, the data acquisition layer is used for acquiring key data from a professional data source, and meteorological data are acquired by means of a professional meteorological API interface, a platform vehicle-mounted ultrasonic meteorological sensor and a vehicle-mounted three-dimensional ultrasonic anemograph; gas concentration data of greenhouse gas is collected by a Picarro analyzer and a PRI analyzer in real time, GPS data is obtained through a vehicle-mounted high-precision RTK multi-frequency Beidou module, and map data is called from a map service API. According to the invention, meteorological and gas concentration multi-source data can be comprehensively collected and integrated, the advanced technology is adopted to efficiently process and analyze the data for real-time collection, real-time analysis and real-time display, the real-time performance is ensured, accurate prediction is realized, various requirements are met, and the operation is convenient; and an advanced architecture and a stable mechanism ensure reliable operation of the system.

The present invention provides a high-speed fault identification method based on Internet of Vehicles big data, comprising the following steps: S1, acquiring vehicle status data, behavior data, and GPS data of an in-vehicle infotainment terminal, transmitting same to the cloud, and preprocessing the vehicle status data, the behavior data, and the GPS data to obtain vehicle speed, sharp acceleration, sharp deceleration, sharp turn, activation status of hazard warning lights, driver exit status, an anti-lock braking system (ABS) fault related trouble code, an engine malfunction indicator lamp related trouble code, a tire pressure anomaly related trouble code, a coolant temperature warning related trouble code, and GPS point positioning data. Thus, whether a vehicle has stopped on a highway due to a fault can be effectively and rapidly detected, and location information of the vehicle is acquired on the basis of the GPS point positioning data, without the need to mount an aftermarket sensor, thereby achieving the advantages of low costs and high accuracy. Multi-dimensional identification of whether the vehicle has stopped on the highway due to the fault provides a basis for an original equipment manufacturer to implement after-sale care and for ensuring driver safety.

The invention provides a radar system error estimation method and system, and relates to the technical field of data processing, and the method comprises the steps: obtaining radar data and GPS data; performing time registration processing on the radar data and the GPS data to obtain time-aligned GPS data; performing filtering preprocessing on the radar data and the time alignment GPS data to obtain filtered radar data and filtered GPS data; according to the filtered GPS data, scaling the filtered radar data through anisotropic scaling based on scale transformation to obtain scaled radar data; carrying out coordinate conversion on the filtered GPS data, and converting the filtered GPS data into a radar polar coordinate system; and in a radar polar coordinate system, registration is carried out on the scaled radar data and the filtered GPS data through an improved iterative nearest point algorithm, and a radar system error is estimated. According to the method, the estimation precision of the radar system error is remarkably improved, and the problem of non-rigid deformation and rigid error coupling is effectively solved.

The invention relates to the technical field of data processing, in particular to a cleaning fusion system for urban traffic multi-source heterogeneous data, which comprises a space-time correlation graph construction module used for creating a unique node identifier for each GPS data record, roadside camera data record and coil data record according to multi-source heterogeneous traffic data, and establishing a spatial adjacent relation edge by judging whether road segment identifiers to which multiple nodes belong are the same or not. According to the method, context awareness evaluation of data source reliability is achieved, meanwhile, when traffic time series data are processed, the data are decomposed into a low-frequency approximation coefficient representing a long-term trend and a high-frequency detail coefficient capturing a sudden change event, and the approximation coefficients of the multi-source data are fused through the obtained context awareness credibility score; meanwhile, a maximum value selection strategy is adopted for detail coefficients, key signal features caused by real traffic events can be reserved, and the fidelity of fused traffic data is improved.

The invention discloses an unmanned vehicle positioning method and system based on multi-sensor fusion, and belongs to the technical field of unmanned driving and robots. The method mainly comprises the following steps: fusing laser radar and IMU data through an LIO module, generating a laser radar odometer, and aligning the laser radar odometer with GPS data through SVD decomposition so as to deduce latitude and longitude information in a GPS-free environment; a deep learning model is introduced to enhance the robustness of label detection through an improved AprilTag visual module, and the precise longitude and latitude of a camera are calculated through nonlinear optimization in combination with IMU attitude data and the prior position of a label; and finally, according to error characteristics of the two modules, performing dynamic weighted fusion on positioning results, and performing optimization by using an error state Kalman filter (ESKF) to obtain a smooth and high-precision final pose. According to the invention, the problem of continuous and accurate positioning of the unmanned vehicle in a GPS signal interruption or complex environment is effectively solved, and the reliability and robustness of the system are improved.

The invention discloses a ship pier anti-collision early warning method and system based on a laser radar, and relates to the field of ship collision early warning, and the method comprises the steps: judging the movement condition of a ship based on GPS data and imu data, and collecting the three-dimensional point cloud data in front of a ship driving route when detecting that the ship moves towards a collision object; converting the three-dimensional point cloud data from a radar coordinate system to a ned coordinate system, and filtering and segmenting the three-dimensional point cloud data after coordinate conversion to obtain a point cloud cluster of the collision object; based on the distance from each point in the point cloud cluster to the ship contour, a preset distance threshold value and a preset number threshold value, executing multi-level collision risk assessment so as to output an early warning signal of a corresponding level; according to the method, the three-dimensional point cloud data is filtered and segmented, noise point clouds are effectively eliminated, the point cloud cluster representing the collision object is accurately extracted, and refined and graded early warning of the collision risk is further realized through a multi-stage risk assessment mechanism.

The invention relates to a degradation environment mapping technology, and discloses a degradation environment environment map construction method comprising the following steps: obtaining current frame IMU data, and carrying out IMU integral calculation on the current frame IMU data and historical frame IMU data to obtain a current frame IMU factor; obtaining GPS data of a current frame, and calculating a GPS factor of the current frame by combining the IMU data of the current frame and utilizing a preset filtering algorithm; acquiring current frame laser radar data, and performing feature extraction on the current frame laser radar data to obtain laser radar features; performing degradation detection according to the laser radar features and then performing point cloud registration to obtain laser radar factors; and performing factor graph optimization on the laser radar factor, the current frame IMU factor and the current frame GPS factor by using a factor graph optimization algorithm to obtain a degraded environment map. The invention further provides an environment map construction device of the degraded environment, electronic equipment and a storage medium. According to the invention, the accuracy of mapping in the degraded environment can be improved.

The invention provides an intelligent scene classification method and device based on a vehicle track, equipment and a medium. The problems that an existing intelligent driving scene classification labeling method is low in labeling efficiency and cannot give full play to the advantages of track data are effectively solved. The method comprises the following steps: starting a micro-service for scene intelligent classification of the vehicle; the micro-service downloads mcap data according to an mcap data list corresponding to the data updating information, extracts GPS data and IMU data in the mcap data, and combines the GPS data and the IMU data to obtain an initial vehicle track; intercepting a target vehicle track from the initial vehicle track, extracting vehicle state information and road information in the target vehicle track, and labeling corresponding labels; and integrating the labels of the multiple segments of target vehicle tracks to classify a target scene where the vehicle is located, and performing intelligent classification on the scene of the vehicle based on the target scene.

The invention relates to a digital intelligent trajectory prediction method and system based on logistics transportation order tracking. The method comprises the steps of collecting multi-source transportation data including vehicle GPS data, AI telephone interaction data, ETC passing record data and freight station transfer operation node data; performing space-time alignment, data cleaning and map matching based on road network information on the multi-source transportation data; utilizing the processed multi-source data to generate a continuous digital intelligent track through an interpolation algorithm and a path deduction method; calculating the latest predicted position of the vehicle through a prediction algorithm model based on the digital intelligent track in combination with real-time traffic information and a historical driving mode; residual paths, real-time road conditions, weather factors and driver behavior modes are comprehensively considered, and the predicted arrival time is calculated through a dynamic ETA model. According to the invention, the GPS signal blind area problem is effectively solved, the whole-course blind-area-free tracking and high-precision arrival time prediction are realized, and the transparency and intelligent management level of logistics transportation are greatly improved.

The application is suitable for the field of spatio-temporal data mining technology, and provides a taxi supply and demand imbalance road section detection method, which comprises the following steps: extracting a GPS data entry sequence corresponding to each license plate from a taxi GPS original data set in chronological order; determining a taxi supply point and a taxi demand point according to all GPS data entry sequences; determining a correction point of the taxi supply point and the taxi demand point in a city road network; finely dividing the city road network to obtain a final city road network; taking a road section with a taxi demand correction point in the final city road network as a seed road section, and expanding the seed road section along the final city road network through an iterative optimization method to obtain a plurality of candidate taxi supply and demand imbalance road sections; performing significance test by using a Monte Carlo simulation method, and taking a candidate taxi supply and demand imbalance road section that passes the significance test as a taxi supply and demand imbalance road section. The application can improve the accuracy of taxi supply and demand imbalance road section detection.

To provide a system for preventing an accident by detecting a recognition discrepancy or an operation mistake between a controller and a pilot in real time.SOLUTION: A system comprising: means for receiving voice communication data between an air traffic controller and a pilot of an aircraft; means for converting the received voice communication data into text data; means for extracting instruction content from the converted text data; means for generating a predicted action based on the extracted instruction content; means for collecting current position data of the aircraft from a GPS; means for monitoring an actual action of the aircraft based on the collected GPS data; means for comparing the monitored action of the aircraft with the predicted action to detect an abnormality; and means for generating and notifying an alert when the abnormality is detected.SELECTED DRAWING: Figure 1

The invention provides a traffic participant road condition early warning method and system based on a smart phone, and relates to the technical field of active safety. The method comprises the following steps: acquiring a video stream through a smart phone camera, and synchronously acquiring data of a gyroscope, an accelerometer, an ambient light sensor and a GPS; identifying a user motion state and an environment scene based on the sensor data; carrying out target detection, semantic segmentation and distance estimation on the video frame by adopting a lightweight neural network; performing risk assessment according to the target type, the distance and the movement speed of the user, and generating an early warning instruction; a multi-mode warning is sent to a user in a voice, vibration, vision or interface switching mode; and dynamically adjusting the analysis frequency and the power consumption mode according to the environmental complexity. Existing hardware of the smart phone is fully utilized, extra equipment is not needed, low-cost and high-universality real-time environment perception and active early warning are achieved, and the travel safety of traffic participants such as walking and riding is remarkably improved.