Data storage and retrospective platform for n3 sweepers

Abstract

The invention discloses a data storage and backtracking platform for N3 electric sweepers, which consists of two parts: an online end and an offline end; Bodywork gear operation, providing data for the bodywork gear selection module, and providing the best bodywork gear selection under the current garbage scene for the online vehicle speed optimization control module, so as to realize the decoupling of bodywork operation and vehicle speed control; the offline terminal uses the online terminal The collected data is trained for the bodywork gear selection algorithm, and the relevant parameters in the bodywork gear selection algorithm are obtained and sent to the online bodywork gear selection module, which is used to iteratively update the online bodywork gear control algorithm, and clean the The neural network parameters in the vehicle's online speed optimization control module are fused and transmitted to each sweeping vehicle; the online and offline terminals realize data transmission through the control information update system.

Description

Technical field [0001] The present invention relates to the field of sweeping cars, which specifically relates to data storage and retrospective platforms facing the N3 type cleaning vehicle. Background technique [0002] At present, to achieve the intelligence of the N3-type pure electric sweeper road operation, it is necessary: ​​(1) Power matching test in advance, the artificial arrangement scene will select the best gear to obtain power matching test data; (2) Use these power matches Test data training out power matching model to simulate the driver's operational behavior; (3) When the actual cleaning operation, the category of the current scene road garbage is output based on the visual sensor and the perceptual algorithm, input to the coverage information, input to A model of power matching test data is trained, thereby automatically outputting the upload system gear of the current scene demand. [0003] This technical route for such sweeping vehicles has the following issu...

AI Technical Summary

Problems solved by technology

[0003] The technical route of the intelligent operation of this sweeping vehicle has the following problems: (1) The number and types of artificially arranged scenes cannot truly reflect the complex and diverse road conditions, resulting in the data of the power matching test cannot truly reflect the actual road surface, thus affecting The precision and accuracy of the trained power matching model; (2) Each power matching test needs to set up the scene first and then select the best gear for operation, which means that obtaining power matching data requires a lot of manpower, material resources and Time cost, which also makes the power matching model unable to be continuously updated

Images