Shulou(Shulou.com)05/31 Report--

This article is about how Apollo adds a new evaluator to the prediction module. The editor thinks it is very practical, so share it with you as a reference and follow the editor to have a look.

Brief introduction

The evaluator generates features (original information from obstacles and current vehicles) by applying a pre-trained deep learning model to obtain model output.

To add an evaluator

Follow these steps to add an evaluator named NewEvaluator:

Add a field to the proto

Declare a class NewEvaluator that inherits from the Evaluator class

Implementation class NewEvaluator

Update Forecast configuration

Update evaluator management

Let's use the above method to add a new evaluator.

Declare a class that inherits from the Evaluator class

NewEvaluator

Create a new file new_evaluator.h under the modules/prediction/evaluator/vehicle directory. The statement is as follows:

# include "modules/prediction/evaluator/evaluator.h" namespace apollo {namespace prediction {class NewEvaluator: public Evaluator {public: NewEvaluator (); virtual ~ NewEvaluator (); void Evaluate (Obstacle* obstacle_ptr) override; / / Other useful functions and fields.} / / namespace prediction} / / namespace apollo

2. Implementation class NewEvaluator

Create a new file new_evaluator.cc in the same directory as new_evaluator.h. The implementation is as follows:

# include "modules/prediction/evaluator/vehicle/new_evaluator.h" namespace apollo {namespace prediction {NewEvaluator::NewEvaluator () {/ / Implement} NewEvaluator::~NewEvaluator () {/ / Implement} NewEvaluator::Evaluate (Obstacle* obstacle_ptr) () {/ / Extract features / / Compute new_output by applying pre-trained model} / / Other functions} / / namespace prediction} / / namespace apollo

Add a new evaluator to proto

Add a new evaluator type to prediction_conf.proto:

Enum EvaluatorType {MLP_EVALUATOR = 0; NEW_EVALUATOR = 1;}

Update the prediction_conf file

In modules/prediction/conf/prediction_conf.pb.txt, update the field evaluator_type as follows:

Obstacle_conf {obstacle_type: VEHICLE obstacle_status: ON_LANE evaluator_type: NEW_EVALUATOR predictor_type: NEW_PREDICTOR}

Update the management of the evaluator

Update CreateEvluator (...) as follows:

Case ObstacleConf::NEW_EVALUATOR: {evaluator_ptr.reset (new NewEvaluator ()); break;}

Update RegisterEvaluators () as follows:

RegisterEvaluator (ObstacleConf::NEW_EVALUATOR)

After completing the above steps, the new evaluator is created successfully.

Add a new feature

If you want to add new features, please follow these steps:

Add a field to the proto

Suppose the name of the new evaluation result is new_output and the type is int32. If the output is directly related to the obstacle, you can add it to the modules/prediction/proto/feature.proto, as follows:

Message Feature {/ / Other existing features optional int32 new_output = 1000;}

If the output is lane-related, add it to the modules/prediction/proto/lane_graph.proto as follows:

Message LaneSequence {/ / Other existing features optional int32 new_output = 1000;} Thank you for reading! This is the end of the article on "how to add a new evaluator to the prediction module of Apollo". I hope the above content can be of some help to you, so that you can learn more knowledge. if you think the article is good, you can share it for more people to see!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.