# US 2022/0035015 A1 · Method and system for processing obstacle detection result of ultrasonic sensor array > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2022/0035015 A1 - Document ID: US20220035015A1 - Status: Published application - Kind: A1 - Application number: 17/278,248 - Filed: 2019-12-18 - Published: 2022-02-03 - Earliest claimed: 2019-01-15 - Assignees: Beijing Baidu Netcom Science Technology Co., Ltd. - Applicants: Beijing Baidu Netcom Science and Technology Co., Ltd. - Inventors: Xiaoxing Zhu; Xiang Liu; Fan Yang - Web page: https://patentgrep.com/patent/US20220035015A1 ## CPC - G01S 7/52004: {Means for monitoring or calibrating (short-range imaging G01S7/5205)} - G01S 15/42: Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S15/46) - G01S 15/86: Combinations of sonar systems with lidar systems; Combinations of sonar systems with systems not using wave reflection - G01S 15/931: of land vehicles - G01S 17/42: Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S17/46) - G01S 17/931: of land vehicles - G06N 3/044: Recurrent networks, e.g. Hopfield networks - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/045: Combinations of networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G01S 2015/938: {in the bumper area} - G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract The present disclosure provides a method and system for processing an obstacle detection result of an ultrasonic sensor array. The method comprises: obtaining obstacle coordinates collected by ultrasonic sensors in an ultrasonic sensor array; inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model, to obtain true or false labels for the obstacle coordinates collected by the ultrasonic sensors output by the pre-trained neural network model; process the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels. The present disclosure improves the accuracy in the obstacle detection of the ultrasonic sensor array, avoids detection error and omission and improves the driving safety. ## Claims ### Claim 1 A method for processing an obstacle detection result of an ultrasonic sensor array, wherein the method comprises: obtaining obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array; inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model, to obtain true or false labels for the obstacle coordinates collected by the ultrasonic sensors output by the pre-trained neural network model; processing the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels. ### Claim 2 (depends on 1) The method according to claim 1, wherein the inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model comprises: selecting feature values of to-be-processed obstacle coordinates collected by an ultrasonic sensor and its adjacent N ultrasonic sensors to generate a data sequence, where N is a positive integer smaller than the total number of the ultrasonic sensors in the ultrasonic sensor array; inputting the data sequence to the pre-trained neural network model. ### Claim 3 (depends on 1) The method according to claim 1, wherein the neural network model is trained by the following steps: building an obstacle test scenario; obtaining and unifying the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array and the coordinates of the same obstacle collected by a LiDAR into a reference coordinate system; performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR, and generating training samples; training the neural network model according to the training samples. ### Claim 4 (depends on 3) The method according to claim 3, wherein the performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR comprises: if an error between the obstacle coordinates collected by the LiDAR and the obstacle coordinates collected by the ultrasonic sensors is within a preset threshold range, labeling the obstacle coordinates as true; if the error is greater than the preset threshold range, labeling the obstacle coordinates as false. ### Claim 5 (depends on 3) The method according to claim 3, wherein the generating training samples comprises: generating the training samples according to the feature values of to-be-processed obstacle coordinates collected by an ultrasonic sensor and its adjacent N ultrasonic sensors and the true or false labels for the to-be-processed obstacle coordinates collected by the ultrasonic sensors, where N is a positive integer smaller than the total number of the ultrasonic sensors in the ultrasonic sensor array. ### Claim 6 (depends on 1) The method according to claim 1, wherein the neural network model is a convolutional neural network model. ### Claim 7 (depends on 1) The method according to claim 1, wherein the neural network model is a Long Short-Term Memory neural network model. ### Claim 8 (depends on 2) The method according to claim 2, wherein the processing the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels comprises: if the labeling result output by the neural network model is true, retaining the to-be-processed obstacle coordinates collected by ultrasonic sensors; if the labeling result output by the neural network model is false, discarding the to-be-processed obstacle coordinates collected by ultrasonic sensors. ### Claim 9 An electronic device, comprising: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform a method for processing an obstacle detection result of an ultrasonic sensor array, wherein the method comprises: obtaining obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array; inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model, to obtain true or false labels for the obstacle coordinates collected by the ultrasonic sensors output by the pre-trained neural network model; processing the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels. ### Claim 10 (depends on 9) The electronic device according to claim 9, wherein the inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model comprises: selecting feature values of to-be-processed obstacle coordinates collected by an ultrasonic sensor and its adjacent N ultrasonic sensors to generate a data sequence, where N is a positive integer smaller than the total number of the ultrasonic sensors in the ultrasonic sensor array; inputting the data sequence to the pre-trained neural network model. ### Claim 11 (depends on 9) The electronic device according to claim 9, wherein the neural network model is trained by the following steps: building an obstacle test scenario; obtaining and unify the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array and the coordinates of the same obstacle collected by a LiDAR into a reference coordinate system; performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR, and generating training samples; training the neural network model according to the training samples. ### Claim 12 (depends on 11) The electronic device according to claim 11, wherein the performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR comprises: if an error between the obstacle coordinates collected by the LiDAR and the obstacle coordinates collected by the ultrasonic sensors is within a preset threshold range, labeling the obstacle coordinates as true; if the error is greater than the preset threshold range, labeling the obstacle coordinates as false. ### Claim 13 (depends on 11) The electronic device according to claim 11, wherein the generating training samples comprises: generating the training samples according to the feature values of to-be-processed obstacle coordinates collected by an ultrasonic sensor and its adjacent N ultrasonic sensors and the true or false labels for the to-be-processed obstacle coordinates collected by the ultrasonic sensors, where N is a positive integer smaller than the total number of the ultrasonic sensors in the ultrasonic sensor array. ### Claim 14 (depends on 9) The electronic device according to claim 9, wherein the neural network model is a convolutional neural network model. ### Claim 15 (depends on 9) The electronic device according to claim 9, wherein the neural network model is a Long Short-Term Memory neural network model. ### Claim 16 (depends on 10) The electronic device according to claim 10, wherein the processing the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels comprises: if the labeling result output by the neural network model is true, retain the to-be-processed obstacle coordinates collected by ultrasonic sensors; if the labeling result output by the neural network model is false, discard the to-be-processed obstacle coordinates collected by ultrasonic sensors. ### Claim 17 A non-transitory computer-readable storage medium storing computer instructions therein, wherein the computer instructions are used to cause the computer to perform an obstacle detection result of an ultrasonic sensor array, wherein the method comprises: obtaining obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array; inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model, to obtain true or false labels for the obstacle coordinates collected by the ultrasonic sensors output by the pre-trained neural network model; processing the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array according to the true or false labels. ### Claim 18 (depends on 17) The non-transitory computer-readable storage medium according to claim 17, wherein the inputting the obstacle coordinates collected by ultrasonic sensors in the ultrasonic sensor array into a pre-trained neural network model comprises: selecting feature values of to-be-processed obstacle coordinates collected by an ultrasonic sensor and its adjacent N ultrasonic sensors to generate a data sequence, where N is a positive integer smaller than the total number of the ultrasonic sensors in the ultrasonic sensor array; inputting the data sequence to the pre-trained neural network model. ### Claim 19 (depends on 17) The non-transitory computer-readable storage medium according to claim 17, wherein the neural network model is trained by the following steps: building an obstacle test scenario; obtaining and unifying the obstacle coordinates collected by the ultrasonic sensors in the ultrasonic sensor array and the coordinates of the same obstacle collected by a LiDAR into a reference coordinate system; performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR, and generating training samples; training the neural network model according to the training samples. ### Claim 20 (depends on 19) The non-transitory computer-readable storage medium according to claim 19, wherein the performing true or false labeling for the obstacle coordinates collected by the ultrasonic sensors according to the obstacle coordinates collected by the unified LiDAR comprises: if an error between the obstacle coordinates collected by the LiDAR and the obstacle coordinates collected by the ultrasonic sensors is within a preset threshold range, labeling the obstacle coordinates as true; if the error is greater than the preset threshold range, labeling the obstacle coordinates as false.