TRAIN AUTONOMOUS
PERCEPTION SYSTEM
Product Description
Application including but not limited to: Subway, trams, medium-capacity rail systems, and freight railway.
CiDi addresses the unique needs of the railway transportation industry by applying technology used in Level 4 self-driving commercial vehicles for railway transportation. By utilizing artificial intelligence technology and integrating perception capabilities of LiDAR, camera, and millimeter-wave radars, CiDi introduces the Train Autonomous Perception System (TAPS).
<0.1km/h
VELOCITY SENSING ERROR
500M
CENTIMETER PERCEPTION
Obstacle detection
Obstacle size(m)Maximum perceived distance(m)
Any obstacle
(centimeter perception)
Specific obstacle
3.25×2.6 (such as trains)5001000
1.5×0.4 (pedestrians)200500
0.3×0.3 (box/falling rocks)100250
TRAIN SPEED:
Error＜0.1km/h
TRAIN LOCALIZATION:
Sub-second initial localization, platform positioning error <0.1m, comprehensive track positioning error <0.3m.
Core Function
SENSOR MODULE
*The sensor module can be adjusted according to vehicle structure.
CAMERA LIDAR MILLIMETER-WAVE RADAR
INTELLIGENT PROCESSING HOST
*Industrial AI chip, up to 248 TOPS, supports OTA remote upgrade.
active obstacle
detection
autonomous train
speed measurement
intelligent hierarchical
alarm for intrusion/threats
customizable three-
dimensional clearance limit
fully autonomous
train positioning
independent safety
redundancy
detection of abnormal
displacement of trackside equipment
detection of signals
and turnouts
Sample View
Supports Detection of Any Obstacle Type
Based on our self-developed artificial intelligence algorithms, our solution can accurately identify three-dimensional track areas. The system can intelligently detect any obstacles such as pedestrians, discarded objects, and falling rocks that may affect the safety of train operations when items intrude into track areas. The system can also accurately determine the level of threat posed by these obstacles.
Supports Detection of Any Obstacle Type
Based on our self-developed artificial intelligence algorithms, our solution can accurately identify three-dimensional track areas. The system can intelligently detect any obstacles such as pedestrians, discarded objects, and falling rocks that may affect the safety of train operations when items intrude into track areas. The system can also accurately determine the level of threat posed by these obstacles.
Fusion of High-Resolution Pixels and Laser Point Cloud Depth
Centimeter-Level Precision Sensing and Positioning
Fusion of High-Resolution Pixels and Laser Point Cloud Depth
Centimeter-Level Precision Sensing and Positioning
Around-the-Clock Operation in All Weather Conditions
Effectively Handling Special Scenarios Such as Rain, Snow, and Nighttime
Around-the-Clock Operation in All Weather Conditions
Effectively Handling Special Scenarios Such as Rain, Snow, and Nighttime
Functional Safety and Certifications
SIL4
JSEP
EN50155
GB/T 25119
Scenario Solution
In the context of low-to-medium capacity rail transportation, where there is no mandatory physical separation in shared road scenarios, other road vehicles and pedestrians coexist within the clearance limits. Thus, there is a higher risk of accidents due to longer braking distances and larger train blind spots. By applying commercial vehicle autonomous driving solutions to low-to-medium capacity trains, multiple safety measures can be provided to mitigate these risks.
INTELLIGENT PROCESSING SENSOR MODULE
Autonomous
environment awareness
track-side
perception
train platform
perception
passenger
flow detection
driver
detection
Perception camera, LiDAR, millimeter-wave radar, 360°perception, an alarm is issued when the estimated TTC (time-to-collision) falls below threshold
Adaptive panorama (360 view)
forward collision warning
pedestrian collision warning
traffic control
lane-departure warning
rear collision warning
blind spot monitoring
Linkage with train control system to achieve automatic driving
Automated emergency braking (AEB)
Adaptive Cruise Control (ACC)
Lane-keeping assist system (LKA)
Beyond visual range detection
Blind sports compensation on dangerous section
perception results sharing, vehicle-road-cloud collaboration
(Note: The equipment is installed at the trackside location)
Platform passenger flow detection
Line crossing detection for waiting passengers
Train platform gap detection
Passenger Counting Camera
Real-time Passenger Load Statistics
Station boarding and alighting passenger count
Carriage crowding analysis
Drowsy driving warning
Phone call handling warning
Distraction warning
Not in the driving seat warning
Smoking warning
Blocked lens warning
Contact information
18670708436(WeChat and Phone with the Same Number)
Building A3-A4, Inspection and Testing Special Industrial Park, No. 336, Xueshi Road, YueLu District, Changsha, Hunan Province, China
reserve Obtaining proposal