Reorganize software stack#168
Merged
nathanaschbacher merged 125 commits intodevelfrom Jul 28, 2017
Merged
Conversation
Prior to this commit, the CAN gateway would receive messages on the OBD bus and then repackage them into "chassis state" messages, and also publish heartbeat messages with chassis state information. This commit removes that overhead by instead having the CAN gateway look for specific OBD messages on the OBD bus and then republish any that it receives to the Control CAN bus for further consumption by the modules without any processing or bundling up into "chassis state" reports. This allows the gateway to simplify its role by not doing any processing of OBD messages but instead simply acting as a bridge between the OBD bus and the control bus, and removes the need for it to keep track of timeouts and health information as that can be done by the modules themselves when they look for the specific OBD messages that they care about.
Prior to this commit, there were exponential filters when reading from vehicle sensors because there was a lot of noise from the brake actuator, but now the brake actuator is much less noisy and exponential filters are unnecessary. This commit removes the exponential filters from the modules.
Prior to this commit, the throttle module contained logic that is being moved out of the firmware and will instead be done at a high level on a PC which communicates with the firmware through an API. This commit simplifies the firmware so that it is instead solely responsible for sending out raw values received from above to the DAC, checking for faults and publishing fault reports, and publishing a (reduced) report.
Prior to this commit, the steering module contained logic that is being
moved out of the firmware and will instead be done at a high level on a
PC which communicates with the firmware through an API. This commit
simplifies the firmware so that it is instead solely responsible for
sending out raw values received from above to the DAC, checking for
faults and publishing fault reports, and publishing a (reduced) report.
Prior to this commit, using cmake to build tests would attempt to locate the signal smoothing library, which has now been removed in the firmware. This commit adjusts the cmake file so that it doesn't try to include that library.
Prior to this, the steering command field for enabling or disabling the steering module was still called 'enabled'. This commit renames it to 'enable' to fit in line with the new naming procedures.
Prior to this commit, the steering DAC min/max values for output weren't constrained enough. This commit constrains them to known values pending further testing.
Prior to this commit, the OSCC interface lived in the joystick commander utility folder. This commit moves it into the common folder of OSCC so it can be utilized as the start of a OSCC API.
…/OSCC into maint/reorganize-software-stack
Prior to this commit, reports and commands followed a naming structure like oscc_report_brake_s. This commit changes it to read more like namespacing, i.e. oscc_brake_report_s, which fits better with the new code organization and API.
Prior to this commit, the throttle module used timestamps to keep track of when to check for faults and when timeouts occurred. This commit replaces the timestamp system with a timer system where faults are checked ten times a second. For command timeouts, a flag is cleared every time a command is received and set at the end of the interrupt service routine - if a command is not received between one ISR call and the next (a span of 100ms), then the timeout flag has not been cleared and a timeout is considered to have occurred. For sensor validity checks, they are run once with every ISR call, meaning they're run every 100ms.
Prior to this commit, the steering module used timestamps to keep track of when to check for faults and when timeouts occurred. This commit replaces the timestamp system with a timer system where faults are checked ten times a second. For command timeouts, a flag is cleared every time a command is received and set at the end of the interrupt service routine - if a command is not received between one ISR call and the next (a span of 100ms), then the timeout flag has not been cleared and a timeout is considered to have occurred. For sensor validity checks, they are run once with every ISR call, meaning they're run every 100ms.
Prior to this commit, the brake module used timestamps to keep track of when to check for faults and when timeouts occurred. This commit replaces the timestamp system with a timer system where faults are checked ten times a second. For command timeouts, a flag is cleared every time a command is received and set at the end of the interrupt service routine - if a command is not received between one ISR call and the next (a span of 100ms), then the timeout flag has not been cleared and a timeout is considered to have occurred. For sensor validity checks, they are run once with every ISR call, meaning they're run every 100ms.
Prior to this commit, brake reports were following the old naming style. This commit updates them. This commit also provides a stubbed out API.
…/OSCC into maint/reorganize-software-stack
Prior to this commit, the brake module contained logic that is being moved out of the firmware and will instead be done at a high level on a PC which communicates with the firmware through an API. This commit simplifies the firmware so that it is instead solely responsible for actuating solenoids, checking for faults and publishing fault reports, and publishing a (reduced) report.
Prior to this commit, the CAN Gateway only forwarded three known OBD frames to the control CAN bus. This commit modifies the gatway to forward all OBD frames to the control CAN bus for use further up the stack.
Prior to this commit, the documentation for joystick commander lived in the OSCC README. This commit gives it its own README file in preparation for moving to a different repo. It also updates the OSCC README to reflect the new changes made to the firmware and API. It also adds some inline comments to the joystick commander example code to further clarify interactions between joystick commander and the OSCC API.
Prior to this commit, the firmware had a concept of what vehicle it was building for which is no longer in line with the way the software stack is architected (i.e., firmware should be consistent across vehicles, all that changes is inside of the vehicle specific header like "kia_soul.h"). This commit removes the KIA Soul references and keeps the firmware vehicle agnostic.
…/oscc into maint/reorganize-software-stack
…PolySync/OSCC into maint/reorganize-software-stack
Prior to this commit, the timers would interrupt the modules while they were talking to the hardware which caused sporadic issues. This commit disable interrupts whenever the firmware is interfacing with a piece of hardware (CAN, sensors, solenoids, etc).
Prior to this commit, joystick commander existed in the main OSCC repo which caused versioning problems where joystick commander might need to be updated but a version bump in the firmware/API was unwanted. This commit removes joystick commander from the main repo, but it lives on in its own separate repo.
Prior to this commit, the property based tests were not using the new directory layout. This commit changes them to find files in the right places.
…PolySync/OSCC into maint/reorganize-software-stack
Prior to this commit, builds were failing because the Jenkinsfile was looking for applications that are no longer part of OSCC. This commit updates them.
…/oscc into maint/reorganize-software-stack
…PolySync/OSCC into maint/reorganize-software-stack
nathanaschbacher
approved these changes
Jul 28, 2017
Contributor
|
+1 nice work. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.