HESAI TECHNOLOGY CO., LTD.

A detection method and system for a LiDAR and a LiDAR are provided. The detection method includes: performing a reflection receiving operation to enable a detector of a channel to be detected to receive multiple echo signals reflected by a target plate, to receive multiple reflectivity measurement values of the channel to be detected for the target plate determined by the LiDAR; and determining a reflectivity parameter of the channel to be detected based on a result of the reflection receiving operation, where the reflectivity parameter of the channel to be detected includes: at least one of reflectivity accuracy of the channel to be detected or reflectivity precision of the channel to be detected. The detection method can determine the consistency of the reflectivity measurement of respective channels of the LiDAR and the reflectivity measurement accuracy of an entire LiDAR system.

A signal transmitting and receiving system and a method thereof are provided in this disclosure. The system includes: a transmitter circuit configured to generate a trigger signal at a trigger moment based on a first clock signal and transmit a detection signal at a transmitting time point; and a receiver circuit configured to receive an echo signal generated by reflection of the detection signal by a target object, and receive the trigger signal based on a second clock signal and enable a first receiving channel at an initial start time point, where the first receiving channel takes the initial start time point as a time origin and senses a return time point of receiving the echo signal: determining a first calculation start point of the first receiving channel, and determining a target time difference between the return time point sensed by the first receiving channel and the transmitting time point based on the first calculation start point, where a first delay of the initial start time point relative to the trigger time point includes an unpredictable jitter delay. Based on the signal transmitting and receiving system and the method thereof of this disclosure, a problem of a jitter of a scanning start point when a receiving system and a transmitting system are clocked asynchronously can be solved, and it is possible to accumulate scanning data in multiple scans in a chip, thereby further increasing a signal-to-noise ratio and improving a ranging capability and application scenarios of a receiving chip.

This disclosure discloses control methods and apparatuses for LiDAR. In an implementation, a method comprises: determining, for a group of channels in the LiDAR, a first light-emitting time sequence and a second light-emitting time sequence for lasers of the group of channels, wherein the first light-emitting time sequence is different from the second light-emitting time sequence, the lasers of the group of channels are configured to emit light in parallel, and each channel of the group of channels comprises a laser and a detector, controlling the lasers of the group of channels to emit a second group of detection pulses based on the second light-emitting time sequence in the second detection orientation, and determining, for a channel in the group of channels, an effective echo pulse of the channel based on the first group of echo pulses and the second group of echo pulses of the channel.

A LiDAR, a detection method thereof, and a readable storage medium are disclosed. The LiDAR includes multiple lasers and multiple detectors to form multiple channels, and the detection method includes: performing detection by group for all channels, where the detection includes: emitting, by a laser, an optical signal, receiving, by a detector, an echo signal reflected by an obstacle, and determining detection information on the obstacle based on the echo signal; emitting, by each group of lasers, light in parallel; a single detection of each channel including a first sub-detection and a second sub-detection; and correspondingly changing a light-emitting strategy of a laser of a parallel light-emitting channel in at least one of the second sub-detection or a first sub-detection of a subsequent round based on an echo signal in the first sub-detection. Solutions of embodiments of this disclosure can reduce crosstalk between parallel light-emitting channels and improve detection quality.

A method for processing detection result includes determining a signal peak of a target pixel signal, and determining an accumulated signal by accumulating the target pixel signal based on a determination that the target pixel signal and neighboring pixel signals satisfy an accumulation condition and the signal peak of the target pixel signal is not greater than a first detection threshold.

A method for processing a point cloud generated by a light detection and ranging system, includes: receiving a first measurement generated by the light detection and ranging system; retrieving a plurality of second measurements that are adjacent to the first measurement; calculating a first parameter by using distances of the first measurement and the plurality of the second measurements, the first parameter indicating a degree of continuum of the first measurement relative to the plurality of the second measurements; and determining whether the first measurement represents a measurement of noises by using the first parameter.

This disclosure provides a solid-state LiDAR and a method for controlling a solid-state LiDAR. The solid-state LiDAR includes: an emitter module and a receiver module, where the emitter module can emit a detection signal, and includes multiple light-emitter units; and the receiver module can receive an echo signal of the detection signal reflected by an obstacle, and includes multiple groups of detectors; and where an emitting sub-field of view corresponding to each light-emitter unit is coincident with a receiving sub-field of view corresponding to at least one group of detectors, and an angular range of the emitting sub-field of view corresponding to each light-emitter unit is greater than an angular range of the receiving sub-field of view corresponding to a coincident group of detectors.

A detection method for a LiDAR is provided. The LiDAR includes multiple detection channels. Each detection channel includes at least one laser and at least one detector. The laser of each detection channel is configured to emit a sequence of detection pulse including at least one detection pulse. The detection method includes: at S 11 , determining whether there is an object within a first distance, based on a previous detection result, wherein the first distance is not greater than 1/10 of a maximum detectable distance of the LiDAR; at S 12 , dynamically changing a code of a sequence of detection pulse of the detection channel, based on the determining result; at S 13 , controlling a laser of the detection channel to emit a detection pulse based on the code; at S 14 , receiving an echo of the detection pulse reflected by the object, and determining information of the object based on the echo. In this disclosure, by analyzing the previous detecting result to predetermine whether an object is continuously detected within a medium or near distance, the code of the sequence of detection pulse is dynamically changed, so that capabilities of distance measurement and anti-interference are both considered.

A drive circuit includes a switch controller, a first monitor, a second monitor, and a safety controller. The switch controller can selectively control a light-emitter apparatus to emit light and stop emitting the light. The first monitor can monitor light-emitting energy of the light-emitter apparatus. The second monitor can monitor a light-emitting duration of the light-emitter apparatus. The safety controller can control the light-emitter apparatus to stop emitting the light by a switch controller in response to at least one of the light-emitting energy reaching a first threshold or the light-emitting time reaching a second threshold.

A method and an apparatus for improving the resolution of a LiDAR. The LiDAR includes a plurality of actual channels, each of which corresponds to one emitter unit at an emitting end and one detector unit at a detecting end. The method includes determining at least one interpolation channel to be generated; separately determining, for each interpolation channel to be generated, one or more associated channels among the plurality of actual channels; determining weights of the one or more associated channels with respect to the interpolation channel; and generating a waveform of the interpolation channel based on waveform information of the one or more associated channels and the weights thereof, and determining point cloud data of the LiDAR based on the waveform information of each actual channel and each interpolation channel. The solution of this disclosure can avoid increasing the hardware load thereon.