Remote Driving
CUSTOMER REQUIREMENTS
Working in high-risk areas with frequent accidents.
The harsh working environment results in an inability to work for long periods, leading to low efficiency.
Production efficiency is affected by the operator's mental state, experience and skills.
The 2D display lacks depth perception and a sense of space, resulting in low efficiency.
Remote operation delay.
SOLUTION

Remote driving perception computing based on MIIVII EVO TX2 GMSL2.

yuancheng_r13_c3.png
Taking advantage of AI, AR, and other technologies, we can obtain a large number of on-site construction photos and build models using the MIIVII EVO series for positioning buckets.
yuancheng_r13_c10.png
During the operation, the operator should manually set an unloading point and can reset the device by pressing a single button after excavation. The MIIVII device can set a path based on the specified positioning point for unloading.
yuancheng_r14_c16.png
Optimize the video transmission structure by leveraging NVIDIA's powerful encoding/decoding functionality, control the delay time within 300 ms, and achieve real-time operation 2,500 km away.
CONTACT US
FRAMEWORK
VALUE
01
The average cycle time of a single bucket is shortened from 30 seconds (using manual remote control) to 18 seconds, achieving 90% efficiency compared to manual operation.
02
Unloading is achieved by pressing one key, resulting in higher operational efficiency, with a cycle time for a single bucket of up to 18 seconds.
03
Delay in operating an engineering vehicle at a distance of 2,500 km is less than 300 ms.
04

Remote driving enables avoidance of dangerous environments.
05
2D image positioning is used to determine the drop point of the bucket.