51 results about "Walking velocity" patented technology

The gait monitor system and method provides various basic gait parameters including step count, cadence, and step duration, in addition to its ability to distinguish between normal, limping and shuffling gait modes, as well as determine falls. Moreover, this gait monitor may be provided with additional sensors, e.g. beam break at the beginning and end of a corridor to estimate average walking velocity (with the distance between the beams known or determined); this enables the calculation of additional gait characteristics such as average step length and average stride length. These parameters can additionally be used to detect various gait anomalies and other diagnostic information.

The present invention relates to a video-based method of intense passenger flow counting and automatic detection of pedestrian speed, belonging to the computer vision technical field. The present invention adopts video capture device and process algorithm. Wherein, the video image capture device captures images through completed circuit television (CCTV ), normally with a camera set over the top of entrance and exit of passage way for real-time capture of video images of passengers in and out. The processor adopts computer vision algorithm to process the captured video images and identify the faces with Adaboost algorithm, with Harr characteristics as input, weak classifier weight combination to form a strong classifier, the strong classifier to form a waterfall-like cascading. Then calculation is triggered. The walking speed calculation of pedestrians is mainly based on tracking of the faces. The tracking starts as the pedestrian faces enter the detection zone and stops as the faces exit the detection zone. The keys to the walking speed calculation of pedestrians are targeting and tracking.

A walking control apparatus and method of a robot. The walking control method include confirming a swing leg and a support leg by judging a walking state of the robot when a walking velocity of the robot and a walking command are received by the robot, generating reference pitch knot points of a hip joint unit of the swing leg based on the walking state and the walking velocity of the robot, generating a target pitch angle trajectory of the hip joint unit of the swing leg using the reference pitch knot points, calculating torques tracking the target pitch angle trajectory, and outputting the torques to the hip joint unit of the swing leg to control the walking velocity of the robot. The walking velocity of the robot is rapidly and easily changed by adjusting at least one of a step length and a step time.

The invention discloses a power line inspection robot control method in a windy environment. A wind velocity sensor is arranged on an inspection robot. Real-time wind velocity vector is measured, and measured data are fed back to an inspection robot control system. The inspection robot control system uses a built-in algorithm, and the wind velocity measured by the wind velocity sensor is sent back to the control system and is subtracted by current walking velocity vector to acquire the exact value of the wind velocity in the current natural environment. The value is sent back to a ground station through a digital radio and is provided for an operator. According to the invention, the risk possibility of the inspection robot in the windy environment can be effectively reduced, and the safety of the inspection robot is improved.

A motorized personal transportation article is described which transports a person by wearing a pair of power-assisted motorized shoes, wherein the shoes provide an increase in a user's walking speed in a forward walking action through supplementary motion of moving or propelling means based on an intended walking speed of the user, such that when a user's intended walking speed changes or when a user intends to substantially decelerate or immobilize, the speed of the supplementary motion can be adjusted before the step is completed. The sole of each of the shoes houses at least one plate coupled to moving or propelling means such as conveyor assemblies, and the moving or propelling means are designed to neutralize forces acting to disrupt the operation of the moving or propelling means during a forward walking action or the balance and comfort of the user.

The invention relates to the field of emergency refuge path planning, and aims to provide a planning method suitable for emergency refuge fastest walking paths in mountainous area complex terrains. According to the method, an A* algorithm is used for carrying out shortest-time emergency risk avoiding path planning on a two-dimensional elevation grid topographic map of a planning area; on the basis, the influence coefficient of the local common ground type on the walking speed of the person is further measured through a unified standard, the influence of the gradient and the ground type on thewalking speed of the person is considered, and the A* algorithm is corrected. And a connecting line between paths with smaller G values formed by the grid with the smallest F value is used as an emergency evacuation route with the shortest time from the starting point to the terminal point. Based on the improvement of the traditional A* algorithm, the influence of the ground type and gradient change on path planning is increased, and the emergency refuge path planning result is more suitable for the practical application situation of complex areas in mountainous areas.

The invention discloses a non-contact active and passive hybrid walking aid robot and a control method thereof. The non-contact active and passive hybrid walking aid robot comprises a pair of lateralnon-contact moving chassis, the lateral non-contact moving chassis are fixed through a connecting plate, and the left non-contact chassis and the connecting plate are provided with four eight-channelshort-distance sensors in total; each sensor is provided with eight supports, and each support is provided with a single-channel laser ranging PCB which is connected through electric signals. The walking speed intention of a user under a non-contact condition is recognized, and the walking speed recognition precision is improved while the recognition hysteresis quality is reduced. When the omni-directional moving chassis is applied to the walking rehabilitation training robot with the omni-directional moving chassis, the walking rehabilitation training robot can move and rotate in any direction without changing the posture of the robot, and the turning radius is not needed; and a solution is provided for non-contact active training and passive training fused omni-directional walking rehabilitation training in an indoor narrow space scene.

The present invention provides a plurality of training modes for an electric walking rehabilitative device, including: constant speed mode, adaptive speed mode, strengthening push/pull mode, weight bearing mode, high frequency disturbance random speed mode, low frequency perturbation variable speed mode, direction control by handle push/pull forces mode, and slope mode, to improve the user's walking speed, muscle strength, balance, speed adjustment, direction control and slope walking.

The present invention relates to a traffic light controller and a method of controlling a traffic light using the same. The traffic light controller according to the present invention includes an input unit configured to receive walking speed information of a pedestrian, a memory in which a program for controlling a traffic light using the walking speed information is stored, and a processor configured to execute the program, wherein the processor sets a lighting maintenance time of a green light of a pedestrian traffic light at a crosswalk using the walking speed information and changes the lighting maintenance time according to a result of monitoring a crossing situation.

The present invention comprises: a main body; an information collecting unit for receiving input of user information; a moving guide unit configured to enable a part of a user's body to be in contact with the main body; a moving unit configured to move the main body; and a control unit which, when sensing that a part of a user's body is in contact with the moving guide unit, controls the moving unit so that the main body moves along a moving route extracted from the user information, wherein the control unit controls the moving unit to allow the main body to move according to a walking speed of the user.

The invention discloses a multi-angle tumble high-risk identification method and system based on skeleton key points, and relates to the technical field of medical image data processing, and the method comprises the steps of collecting a video stream of a normal walking speed of a crowd, and obtaining an original frame image sequence set, the video stream being obtained in one direction; based on a human body posture recognition algorithm, processing each original frame image, and obtaining a skeleton key point and a gait contour map corresponding to each original frame image; performing affine transformation on the skeleton key points to obtain skeleton key points at different angles; acquiring corresponding gait data, joint buckling data and a skeleton feature map based on the coordinates of the skeleton key points at different angles; and establishing a gait recognition network, inputting the obtained data into the gait recognition network for tumble high risk, and obtaining a tumble high risk result and a predicted value related to the tumble high risk result, thereby realizing low-cost, rapid and objective screening and recognition of tumble high risk individuals, and realizing prediction of tumble high risk potential individuals.

A mobility device comprising a motorized shoe to be worn by a user to increase the speed of walking. The motorized shoe has a plurality of wheels, with at least one wheel driven by an electric motor through a geartrain. On onboard controller gathers data from at least one of an inertial measurement unit, an ultrasonic sensor, and a vision system to generate a command speed to the electric motor. A user wearing a pair of the mobility devices, one on each foot, is able to walk with a normal gait, but at an increased speed.

The invention discloses a walking speed adjusting method for a biped robot to simulate the gait of a person. The method comprises steps of double-foot supporting phase adjustment and single-foot supporting phase adjustment, the double-foot supporting phase adjustment refers to adjusting the walking speed by adjusting the duration of a double-foot supporting phase, single-foot support phase adjustment refers to adjusting the walking speed by applying torque to change the track energy relative to the pitching degree of freedom of an ankle joint supported by the single-foot support. Compared witha conventional walking speed switching method, the method does not need to design multiple groups of gaits in advance, and achieves speed adjustment and switching based on the speed feedback throughthe change of gait parameters and the application of torque control. Based on the walking speed adjusting method provided by the invention, the robot can be effectively helped to realize real-time speed switching, so walking flexibility and environmental adaptability of the robot are improved.

Preferred embodiments include receiving input defining dimensions for a workcell and a size and location for one or more objects within the workcell, receiving input defining a walking velocity for an operator within the workcell and one or more tasks to be completed during a workcell cycle, receiving input defining one or more times and locations to complete the one or more tasks within the workcell, calculating a total walking time to complete the one or more tasks during the workcell cycle based upon the size and location of the object(s), the time(s) and location(s) to complete the task(s) and the operator walking velocity, and outputting the total walking time for the workcell cycle and a graphical representation of the workcell including the object(s) and an operator walking path between points within the workcell at which the one or more tasks are completed.

The invention discloses a sarcopenia identification method and device based on machine vision, equipment and a medium, and relates to the technical field of medical image data processing. The sarcopenia identification method comprises the following steps: collecting a video stream of the normal walking speed of a crowd, and obtaining an original frame image sequence set; based on a human body posture identification algorithm, processing each original frame image, and obtaining human body key points and gait contour maps corresponding to each original frame image; performing affine transformation on the human body key points to obtain human body key points at different angles; acquiring corresponding gait data, joint flexion data and skeleton feature maps on the basis of coordinates of the human body key points at different angles; and establishing a gait identification network, inputting the obtained data into the gait identification network for sarcopenia identification, and obtaining a sarcopenia identification result and a prediction value related to the sarcopenia identification result. According to the invention, sarcopenia individuals are quickly and objectively screened and identified at low cost, and prediction of potential sarcopenia individuals is achieved.

The invention discloses a crowd risk degree detection method, device and equipment. The method comprises the steps of obtaining the region data of a target region and the crowd data of the target region at a target moment, and enabling the region data to comprise the area, the ground slope and the ground slope length, and enabling the crowd data to comprise the number of crowds, an included angle between a crowd flow direction and a gradient descending direction and average walking speed in the crowds, according to the region data and the crowd data, determining a crowd risk degree of the target region at the target moment, and when the crowd risk degree meets a preset condition, outputting alarm information. By considering the influence of the ground slope, the ground slope length, the average walking speed in the crowd and the included angle between the crowd flow direction and the slope descending direction on the crowd risk degree, the crowd risk degree detection precision of the crowd can be improved, so that accurate early warning is performed, and the occurrence of stampede events is effectively prevented.

The present invention provides a virtual reality walking device comprising: a flywheel which is driven by a foot step part (110) so as to be rotated; a link unit (130) which is provided between the flywheel and the foot step part and provides a walking trajectory to the foot step part; a head-mounted display which provides a virtual reality image; a disk which is attached to the outer circumference of the flywheel so as to integrally rotate with the flywheel, and has grooves (158a) formed at predetermined intervals; a sensor which senses the grooves of the rotating disk; an electric step motor which is attached to the flywheel; a crank of which one end is connected to the electric step motor such that the length thereof can be adjusted, and the other end is connected to the link unit; and a controller which calculates a walking speed on the basis of the number of grooves sensed by the sensor, and is linked with lower-limb walking of a character in the virtual reality image on the basis of the walking speed so as to control the length of the crank, and thus adjusts the stride length of the foot step part.

A method for preventing and detecting a fall of a user and implementing a system including sensors and alarm, a pair of eyeglasses with hinged stems including rims for supporting glasses, worn by the user. The sensors are set in the hinged stems and the rims of the eyeglasses. The sensors include a triaxial accelerometer, an IR transmitter and an IR receiver of infra-red light both directed to the cornea of the user, and a barometric sensor. The triaxial accelerometer signal is acquired and processed to derive a walking pace parameter, a sit to stand parameter, a head posture parameter and an acceleration magnitude over the three axes. A composite index is computed based on the walking pace, the sit to stand and the head posture parameters. The alarm is generated if the first composite index exceeds a threshold.

An exercise test evaluation system includes an acceleration sensor that is worn by a user during an exercise test and acquires acceleration values of a foot of the user, a heart rate sensor that is worn by the user and measures a heart rate of the user, and a processor. The processor acquires a first maximum acceleration value and a second maximum acceleration value, calculates a reliability score using the first and second maximum acceleration values, acquires a walking velocity and a heart rate of the user and estimates a maximum oxygen intake amount using the user's walking velocity and heart rate if the reliability score is equal to or higher than a specific threshold value, and estimates the maximum oxygen intake amount using the user's walking velocity if the reliability score is lower than the specific threshold value.

The invention discloses a rehabilitation exercise speed calculation model parameter identification method based on a particle swarm optimization algorithm, and relates to a model parameter identification strategy for detecting rehabilitation exercise speed by using portable exercise parameter measurement equipment. Firstly, a walking speed model containing unknown parameters is established; secondly, an optimization algorithm objective function is established; and finally, the target function is optimized by using a particle swarm optimization algorithm to obtain optimal model parameters. Thewrist wearing type exercise measuring equipment is introduced to measure rehabilitation exercise parameters, and compared with foot wearing, convenience is brought to trainees. According to the rehabilitation exercise speed calculation model parameter identification method based on the particle swarm optimization algorithm, a wrist exercise and walking speed relation model is established, and a training speed calculation method is provided for wrist-worn exercise measurement equipment. And meanwhile, model parameter identification is carried out by utilizing a particle swarm optimization algorithm, so that an accurate and rapid model parameter solving method is provided for a wrist movement and walking speed relation model.