ApolloAuto
apollo 9.0 与carla联合仿真存在问题
System information
- **OS Platform and Distribution : Linux Ubuntu 20.04
- **Apollo version : 9.0
- Carla 0.9.15
- 深信科创 Bridge 0.9.14
Steps to reproduce the issue:
在Apollo与Carla联合仿真时，如果bridge中发送车辆状态消息的频率太快（周期<0.1s），就会出现走走停停的现象，即车子给了油门车速刚起来一点就会刹车。设定频率减小后，异常现象会有缓解。推测是因为设定周期大时（或carla反馈数据延迟）vehicle_vel_time>0.1，此时会跳出图2中的逻辑（因此时prev_trajectory.start_point().relative_time=0.1s）；
Supporting materials (screenshots, command lines, code/script snippets):
planning的log文件报错如下：
通过查找原因，发现联合仿真时会进入如下逻辑，然后偶发上述bug的现象；单独apollo pnc模块（不带carla）不会进入该逻辑。
报警函数地址：/modules/planning/planning_base/common/trajectory_stitcher.cc
Apollo9planning_report.pdf
I might have a similar problem, can you please tell me what bridge you are using between Carla and Apollo?
I might have a similar problem, can you please tell me what bridge you are using between Carla and Apollo?
It's from SYNKROTRON. https://github.com/guardstrikelab/carla_apollo_bridge.git
I might have a similar problem, can you please tell me what bridge you are using between Carla and Apollo?
use her bridge，it's ok!
System information
- **OS Platform and Distribution : Linux Ubuntu 20.04
- **Apollo version : 9.0
- Carla 0.9.15
- 深信科创 Bridge 0.9.14
Steps to reproduce the issue:
在Apollo与Carla联合仿真时，如果bridge中发送车辆状态消息的频率太快（周期<0.1s），就会出现走走停停的现象，即车子给了油门车速刚起来一点就会刹车。设定频率减小后，异常现象会有缓解。推测是因为设定周期大时（或carla反馈数据延迟）vehicle_vel_time>0.1，此时会跳出图2中的逻辑（因此时prev_trajectory.start_point().relative_time=0.1s）；
Supporting materials (screenshots, command lines, code/script snippets):
planning的log文件报错如下：
通过查找原因，发现联合仿真时会进入如下逻辑，然后偶发上述bug的现象；单独apollo pnc模块（不带carla）不会进入该逻辑。 报警函数地址：/modules/planning/planning_base/common/trajectory_stitcher.cc
Apollo9planning_report.pdf
Hi, could you clarify what you mean by "high frequency" in Bridge? Are you referring to the fixed_delta_time parameter used to control the simulation speed in Carla, or the update frequency of the planning and control modules in Apollo?
请问您解决这个问题了吗？我使用Apollo 9.0和Carla 0.9.14；发现运行时车辆速度很慢，也出现了这种车速刚起来一点就会刹车的情况。
